No entries match these filters. Clear filters
Goose v1.52.0 adds live voice conversations, Z.AI provider, OpenRouter OpenRouter A unified API gateway that routes requests to dozens of LLM providers under one endpoint, letting cyber tools swap or compare models without rewriting provider-specific integration code. /Jev decision crates, and custom OpenAI base URLs.
└──▷ GET THIS VERSION
copy
$ git clone --branch v1.52.0 https://github.com/aaif-goose/goose.git
$ git checkout v1.52.0
Adds live voice conversations capability in the desktop app for real-time audio interaction Not checked yet Introduces decisions provider crate with OpenRouter and Jev implementations for agent decisioning logic Not checked yet Adds Z.AI Coding Plan provider with streaming tool calls support for code generation workflows Not checked yet Enforces recipe parameter limits (max 32 params, 200 select options, 128 KiB) to constrain recipe inputs Not checked yet Supports OpenAI provider custom base URL configuration in the SDK for proxy or local endpoint routing Not checked yet
+1 more
show less
Adds support for Opus 5.5, GPT-6-sol, and GPT-6-luna models via the canonical registry Not checked yet
6 details in Everything
Goose v1.51.0 adds GPT-live API support, EUrouter declarative provider, and ACP prompt meta routing.
└──▷ GET THIS VERSION
copy
$ git clone --branch v1.51.0 https://github.com/aaif-goose/goose.git
$ git checkout v1.51.0
Adds GPT-live API support for real-time streaming interactions Not checked yet Introduces EUrouter as a declarative provider for gateway pairing Not checked yet Routes Desktop sessions to the state-machine loop via ACP prompt meta Not checked yet Adds operator allowlist configuration for gateway pairing security Not checked yet
4 details in Everything
└──▷ BREAKING ON UPGRADE
! Removes the --planning-mode flag and associated planning functionality from the CLI! Removes the create recipe subcommand entirely from the CLI interface
Goose v1.50 adds GPT-6 Astra models, tool calling tool calling A model's ability to answer with a structured request to run a named function or API, which the application executes and returns. It is how agents take actions. for goose-agent, Databricks AI Gateway path configuration, and AIML API integration.
└──▷ GET THIS VERSION
copy
$ git clone --branch v1.50.0 https://github.com/aaif-goose/goose.git
$ git checkout v1.50.0
└──▷ USE IT
Configure the Databricks AI Gateway path for Kotlin-based integrations to direct requests to a specific gateway endpoint.
shell
copy
export DATABRICKS_AI_GATEWAY_PATH="/v1/chat/completions"
Integrate with the AIML API using the new client library published as a separate npm package.
shell
copy
npm install @goose/ai-ml-client
Supports GPT-6 Astra models Not checked yet Allows Kotlin callers to configure the Databricks AI Gateway path via a new setting or environment variable Not checked yet Integrates with AIML API Not checked yet
3 details in Everything
└──▷ BREAKING ON UPGRADE
! Removes fast model routing functionality! Removes managed model registry from local inference
Goose v1.49.0 adds declarative providers (Opper, Databricks, TrustedRouter, etc.), on_failure hooks, web-search skills, ARM64 ARM64 A 64-bit CPU architecture from ARM Holdings, used in mobile chips, Apple Silicon, and cloud servers; cyber tools target it to run natively on those platforms instead of relying on emulation. desktop, and --with-extension naming.
└──▷ GET THIS VERSION
copy
$ git clone --branch v1.49.0 https://github.com/aaif-goose/goose.git
$ git checkout v1.49.0
└──▷ USE IT
Capture pre-execution failures in custom hooks to handle or log errors before they bubble up.
yaml
copy
# Configure your hook.yaml to include an on_failure block
cat <<EOF > ~/.config/goose/hook.yaml
pre_tool_use:
my_hook:
on_failure: |
echo "Tool failed with exit code ${{exit_code}}" >&2
EOF
Load a specific extension by name during the CLI run without relying on auto-discovery order.
copy
$ goose cli --with-extension my_custom_extension_name
Adds on_failure block for PreToolUse hooks Not checked yet Web-search and browser-use built-in skills Not checked yet Model-native audio transcription provider Not checked yet --with-extension can now name its extension Not checked yet OpenRouter OpenRouter A unified API gateway that routes requests to dozens of LLM providers under one endpoint, letting cyber tools swap or compare models without rewriting provider-specific integration code. session_id forwarding and app category header Not checked yet
+15 more
show less
ACp thinking effort derivation and forwarding Not checked yet GPT-5.6 follow-up support for Codex and Responses API Not checked yet Pre-registered OAuth clients for streamable_http extensions Not checked yet OTEL OTEL An open standard and set of tools for collecting traces, metrics and logs from applications. It lets services export observability data without tying instrumentation to one monitoring vendor. request params, response metadata, tool call parity, and agent identification Not checked yet Opper as a declarative OpenAI-compatible provider Not checked yet Databricks Unity Catalog model services discovery Not checked yet TrustedRouter declarative provider Not checked yet OpenCode Zen gateway as a declarative provider Not checked yet Gondola as declarative OpenAI-compatible provider Not checked yet SayGM as declarative OpenAI-compatible provider Not checked yet Lynkr as declarative OpenAI-compatible provider Not checked yet PleumRouter declarative provider Not checked yet Z.ai GLM-5.3 (replacing GLM-5.2) Not checked yet OpenAI GPT-5.6 (sol/terra/luna) via Bedrock Not checked yet Custom provider cost fields for cost tracking Not checked yet
20 details in Everything
Goose v1.48.0 adds declarative providers, PreToolUseResult hooks, /new CLI command, web-search skill, and OpenTelemetry OpenTelemetry An open standard and set of tools for collecting traces, metrics and logs from applications. It lets services export observability data without tying instrumentation to one monitoring vendor. metadata.
└──▷ GET THIS VERSION
copy
$ git clone --branch v1.48.0 https://github.com/aaif-goose/goose.git
$ git checkout v1.48.0
Adds on_failure block for PreToolUse hooks to capture tool execution outcomes Not checked yet Introduces PreToolUseResult event with stable tool_call_id across the tool lifecycle Not checked yet Adds /new CLI command to start a fresh session without restarting the agent Not checked yet Expands --with-extension CLI flag to accept an extension name directly Not checked yet Includes built-in Web-search and browser-use skills Not checked yet
+16 more
show less
Supports pre-registered OAuth clients for streamable_http extensions Not checked yet Derives and forwards thinking effort from the ACP harness Not checked yet Titles new sessions via _meta.sessionTitle in the ACP harness Not checked yet Adds _goose/unstable/session/extensions/apply method to the ACP interface Not checked yet Exposes OpenTelemetry request parameters, response metadata, and agent identification Not checked yet Forwards session_id and app category header to OpenRouter OpenRouter A unified API gateway that routes requests to dozens of LLM providers under one endpoint, letting cyber tools swap or compare models without rewriting provider-specific integration code. via telemetry Not checked yet Replaces Z.ai GLM-5.2 with GLM-5.3 support for custom provider cost tracking Not checked yet Adds model-native audio transcription capability to the provider system Not checked yet Enables routing for OpenAI GPT-5.6 (sol/terra/luna) via AWS Bedrock Not checked yet Provides GPT-5.6 follow-up support for Codex and Responses API endpoints Not checked yet Adds Select saved recipes when creating a schedule in the Desktop UI Not checked yet Sorts configured providers to the top of the provider list in the Desktop UI Not checked yet Collapses scheduled job sessions into accordion view in chat history Not checked yet Auto-focusses chat input when user starts typing in the Desktop client Not checked yet Shows interactive git branch indicator in the chat bottom bar Not checked yet Displays recent used models in the chat footer model picker Not checked yet
21 details in Everything
Goose v1.47 adds interactive git branch indicator, pre-registered OAuth for streamable_http, and recent model picker
└──▷ GET THIS VERSION
copy
$ git clone --branch v1.47.0 https://github.com/aaif-goose/goose.git
$ git checkout v1.47.0
Adds an interactive git branch indicator in the chat bottom bar Not checked yet Supports pre-registered OAuth clients for streamable_http extensions Not checked yet Shows recently used models in the chat footer model picker Not checked yet
3 details in Everything
Goose v1.46.0 adds hooks with PreToolUse denial, 25+ new providers, goose review, major CLI additions, and broad env-var configurability.
└──▷ GET THIS VERSION
copy
$ git clone --branch v1.46.0 https://github.com/aaif-goose/goose.git
$ git checkout v1.46.0
└──▷ TRY IT
Block a specific tool from executing during a session by wiring a PreToolUse hook that denies it — useful for enforcing policy on sensitive operations.
copy
$ GOOSE_MAX_TOOL_RESPONSE_SIZE=65536 goose session
Pin a stable OAuth callback port so OAuth redirect URIs stay consistent across restarts, useful in environments with strict firewall rules.
copy
$ GOOSE_OAUTH_CALLBACK_PORT=9876 goose session
Run a local code review against the current working directory, delegating checks via the default orchestrator.
Set the transcription language for local Whisper so voice input is recognized correctly without auto-detection overhead.
copy
$ LOCAL_WHISPER_LANGUAGE=fr goose session
Adds LOCAL_WHISPER_LANGUAGE environment variable to configure language for multilingual local voice transcription. Not checked yet Adds GOOSE_MAX_TOOL_RESPONSE_SIZE environment variable to configure the maximum size of tool output returned to the agent. Not checked yet Adds MAX_CODE_BLOCK_LINES environment variable to cap the number of lines rendered in code blocks. Not checked yet Adds GOOSE_FAST_MODEL environment variable (consumed by ModelConfig::with_fast) to designate a fast model for lightweight tasks. Not checked yet Adds GOOSE_DOCS_ROOT environment variable to redirect documentation lookups for air-gapped deployments. Not checked yet
+41 more
show less
Adds GOOSE_OAUTH_CALLBACK_PORT environment variable to pin a stable OAuth redirect_uri port across sessions. Not checked yet Adds AZURE_OPENAI_AD_TOKEN environment variable to support Azure AD (Entra ID Entra ID Microsoft's cloud-based identity and access management service, formerly Azure Active Directory, used by cyber tools to authenticate users and enforce access policies in Microsoft 365 and Azure environments. ) bearer-token authentication for the Azure OpenAI provider. Not checked yet Adds _meta.hidden field to the ACP session/new request to create hidden (background) sessions. Not checked yet Adds Hooks framework with PreToolUse denial, letting operators intercept and block tool calls before execution. Not checked yet Adds --edit flag to the session command to edit conversation history before forking a session. Not checked yet Adds /goal slash command for agent self-evaluation before a session concludes. Not checked yet Adds /status slash command to report current session state. Not checked yet Adds /model slash command to switch the active model mid-session. Not checked yet Adds goose review command for local code review workflows. Not checked yet Adds tab completion and provider switching to the model command. Not checked yet Loads global agent hints from ~/.agents/AGENTS.md automatically each session. Not checked yet Adds Markdown as a format option for session export. Not checked yet Surfaces per-message token count, cost, time-to-first-token time-to-first-token Time to first token, the wait between submitting a request and receiving the first generated token. It includes delays such as queuing and processing the prompt. (TTFT TTFT Time to first token, the wait between submitting a request and receiving the first generated token. It includes delays such as queuing and processing the prompt. ), and tokens-per-second stats in the UI. Not checked yet Tracks cache tokens separately for accurate cost reporting. Not checked yet Streams shell command output to the UI in real time as commands execute. Not checked yet Adds TLS support for acp serve. Not checked yet Adds a resizable sidebar with a drag handle and persisted width in the desktop UI. Not checked yet Adds chat history search in the navigation panel. Not checked yet Adds a search filter to the provider selection grid. Not checked yet Adds a TUI TUI A text-based user interface that runs inside a terminal, using characters and ANSI escape codes to render menus, panes, and widgets without a graphical desktop, letting cyber tools ship rich interactivity with no GUI dependency. diff viewer. Not checked yet Emits GenAI semantic convention attributes in OpenTelemetry OpenTelemetry An open standard and set of tools for collecting traces, metrics and logs from applications. It lets services export observability data without tying instrumentation to one monitoring vendor. (OTEL OTEL An open standard and set of tools for collecting traces, metrics and logs from applications. It lets services export observability data without tying instrumentation to one monitoring vendor. ) spans. Not checked yet Forwards images and MCP MCP Model Context Protocol, a standard for connecting AI applications to external tools and data sources. A compatible server exposes capabilities an assistant can use through a common interface. embedded-resource blobs in Anthropic and Google provider formats. Not checked yet Adds logo_uri to OAuth client ID metadata documents (CIMD CIMD A text-based Short Message Service protocol from Nokia for exchanging SMS messages between applications and SMS centers, used to build tools that send, receive, or test carrier messaging flows. ) so authorization servers can display the goose logo on consent screens. Not checked yet Adds proactive OAuth token refresh to avoid re-authentication on every session. Not checked yet Adds goose://new-session and goose://resume deep links, including support for an initial prompt via goose://new-session. Not checked yet Adds encrypted Nostr session sharing and session import. Not checked yet Adds Hugging Face OAuth support with a dedicated auth tab in settings. Not checked yet Supports MLX MLX Apple's machine-learning framework for Apple silicon, which uses the unified memory shared by CPU and GPU to run and fine-tune models on a Mac. models in the local inference provider. Not checked yet Adds Linux Vulkan support for local inference. Not checked yet Adds a UniFFI SDK for cross-language bindings, Python wheel publishing, and Kotlin FFI with Maven publish. Not checked yet Adds a quarterly recurrence option to the session scheduler. Not checked yet Adds Termux detection in the installer to automatically select the musl portable build. Not checked yet Binds MCP apps to trusted ownership metadata to restrict which extensions can be loaded. Not checked yet Adds a CLI command to list available skills with their token counts. Not checked yet Adds an image read tool for ingesting images directly in sessions. Not checked yet Adds Summon extension peek mode for async background tasks, plus context and working_dir parameters for delegate tasks. Not checked yet Adds new providers: Celeris, Friendli, Azure AI Foundry (multi-LLM LLM A large language model, a neural network trained on large amounts of text to recognize patterns and generate language. Applications use it for tasks such as answering, summarizing and writing code. ), OllamaCloud (dynamic model discovery), Sakana AI (Fugu API), iFlytek Spark, Astron MaaS, Fireworks AI, Together AI, OrcaRouter, Perplexity, Alibaba/Qwen (DashScope), Databricks AI Gateway, Scaleway, NEAR AI Cloud, EmpirioLabs, xAI SuperGrok (OAuth subscription), Vercel AI Gateway, Routstr, FuturMix, oMLX, Atomic Chat, Muse Spark 1.1 (Meta Models API), and AWS Bedrock support for OpenAI GPT models via mantle endpoint. Not checked yet Adds support for OpenRouter OpenRouter A unified API gateway that routes requests to dozens of LLM providers under one endpoint, letting cyber tools swap or compare models without rewriting provider-specific integration code. request parameters. Not checked yet Adds new models: GPT-5.6, GPT-5.5, GLM-5.2, Opus5 (with adaptive thinking), claude-sonnet-5, claude-fable-5, Cursor composer-2.5, Cursor-agent (fetched from CLI), latest Gemini models, Gemini 3.x, MiniMax-M3/M2.7, and OVHcloud Qwen3.6-27B. Not checked yet Adds desktop localizations for French, German, Italian, Portuguese, Indonesian, Malay, Vietnamese, zh-TW, Korean, Spanish, Japanese, Hindi, Russian, Turkish, Simplified Chinese (zh-CN), and language selection in the UI. Not checked yet Adds structured summary output with template rendering for context compaction context compaction Condensing an AI conversation or working history to free space in the model's context window. A summary preserves useful information, but details can be lost. . Not checked yet
46 details in Everything
└──▷ BREAKING ON UPGRADE
! The goose project command (and its p alias) has been removed; CLI project support is no longer available.
Goose adds offline documentation skill and Azure AI Foundry integration for OpenAI, Anthropic, and partner models.
New goose-doc-guide skill reads official Goose documentation before answering, enabling offline/air-gapped doc lookups. Not checked yet Adds Azure AI Foundry integration, supporting OpenAI, Anthropic, and partner model deployments from an Azure AI Foundry project. Not checked yet
2 details in Everything
Goose v1.45.0 adds tab completion for model switching, air-gapped docs support, disabling built-in skills, and Gemini/Opus5 model support.
└──▷ GET THIS VERSION
copy
$ git clone --branch v1.45.0 https://github.com/aaif-goose/goose.git
$ git checkout v1.45.0
└──▷ TRY IT
Point Goose at a local mirror of the docs in an air-gapped environment so help content loads without internet access.
copy
$ GOOSE_DOCS_ROOT=/mnt/internal/goose-docs goose session
Run the ACP server with scheduled recipe execution enabled (opt-in after the new default disables it).
copy
$ goose server --enable-scheduler
Adds tab completion and inline provider switching to the model command. Not checked yet Supports latest Gemini models and Opus5 (including adaptive thinking). Not checked yet Adds GOOSE_DOCS_ROOT environment variable for air-gapped / offline documentation access. Not checked yet Gates tool-call label enrichment on ACP client capability, improving ACP interoperability. Not checked yet Adds stable identity for agent event messages, enabling reliable message tracking across sessions. Not checked yet
5 details in Everything
└──▷ BREAKING ON UPGRADE
! The ACP scheduler is now disabled by default; deployments relying on scheduled recipe execution must explicitly pass --enable-scheduler.
Goose adds rp-why skill: an AI collaboration coach that scores your tool sophistication and prompt depth.
└──▷ TRY IT
Install the rp-why skill and initialize your baseline before your first analysis.
copy
$ npx skills add https://github.com/block/agent-skills --skill rp-why
/rp-why init
At the end of a Goose session, check whether you stretched or coasted compared to your typical patterns.
On Friday, compare the week's sessions against your baseline to identify next week's focus areas.
New rp-why skill analyzes conversation history to score sessions on Gas Town Stage (tool sophistication, stages 1–8) and DOK Level (prompt cognitive depth, levels 1–4). Not checked yet New /rp-why init command establishes a personal baseline by analyzing available conversation history (~30 seconds). Not checked yet New /rp-why current command shows how the active session compares to your typical usage patterns. Not checked yet New /rp-why compare command diffs the current session against your baseline to surface progress or regression. Not checked yet Session output includes a quadrant map (Underutilizing / Frontier / Learning Zone / Thinking Ahead), DOK distribution histogram, and actionable growth nudges. Not checked yet
5 details in Everything
Goose adds rp-why skill: an AI collaboration coach that scores your tool sophistication and prompt depth.
└──▷ TRY IT
Install the rp-why skill and initialize your baseline before your first analysis.
copy
$ npx skills add https://github.com/block/agent-skills --skill rp-why
/rp-why init
At the end of a Goose session, check whether you stretched or coasted compared to your typical patterns.
On Friday, compare the week's sessions against your baseline to identify next week's focus areas.
New rp-why skill analyzes conversation history to score sessions on Gas Town Stage (tool sophistication, stages 1–8) and DOK Level (prompt cognitive depth, levels 1–4). Not checked yet New /rp-why init command establishes a personal baseline by analyzing available conversation history (~30 seconds). Not checked yet New /rp-why current command shows how the active session compares to your typical usage patterns. Not checked yet New /rp-why compare command diffs the current session against your baseline to surface progress or regression. Not checked yet Session output includes a quadrant map (Underutilizing / Frontier / Learning Zone / Thinking Ahead), DOK distribution histogram, and actionable growth nudges. Not checked yet
5 details in Everything
Goose v1.44.0 adds new model support (GPT-5.6, Gemini, Muse Spark 1.1), new providers, --edit session flag, per-message usage stats, and OllamaCloud integration.
└──▷ GET THIS VERSION
copy
$ git clone --branch v1.44.0 https://github.com/aaif-goose/goose.git
$ git checkout v1.44.0
└──▷ TRY IT
Edit a prior session's conversation before forking it — useful for pruning context or replaying with adjusted prompts.
copy
$ goose session --edit
Adds support for latest Gemini models. Not checked yet Adds Muse Spark 1.1 support via the Meta Models API. Not checked yet Adds MiniMax-M3 and missing M2.7 model variants. Not checked yet Adds declarative Sakana AI provider for the OpenAI-compatible Fugu API. Not checked yet Adds iFlytek Spark and Astron MaaS providers. Not checked yet
+10 more
show less
Adds Fireworks AI as a declarative provider. Not checked yet Adds OllamaCloudProvider with dynamic model discovery and context limits. Not checked yet Adds OpenRouter OpenRouter A unified API gateway that routes requests to dozens of LLM providers under one endpoint, letting cyber tools swap or compare models without rewriting provider-specific integration code. request parameters support. Not checked yet New --edit session flag to edit a conversation before forking it. Not checked yet New per-message usage stats UI showing tokens, cost, TTFT TTFT Time to first token, the wait between submitting a request and receiving the first generated token. It includes delays such as queuing and processing the prompt. , and tokens/second. Not checked yet Chunks command-classifier input with overlapping windows for improved security robustness. Not checked yet Adds search filter for the provider grid in the UI. Not checked yet Adds delete support for custom apps from the Apps UI. Not checked yet Reconnects desktop ACP sessions automatically after sleep or connection loss. Not checked yet Adds French, German, Italian, Portuguese, Indonesian, Malay, Vietnamese, and zh-TW desktop locales. Not checked yet
15 details in Everything
Goose v1.43.0 adds per-message cost/token stats, OllamaCloud provider, --edit session flag, TLS for ACP serve, and six new model providers.
└──▷ GET THIS VERSION
copy
$ git clone --branch v1.43.0 https://github.com/aaif-goose/goose.git
$ git checkout v1.43.0
└──▷ TRY IT
Review and trim a session's conversation before branching it into a new task, avoiding stale context.
copy
$ goose session --edit
Adds --edit session flag to edit conversation history before forking a session. Not checked yet New per-message usage stats UI showing tokens, cost, TTFT TTFT Time to first token, the wait between submitting a request and receiving the first generated token. It includes delays such as queuing and processing the prompt. , and tok/s for each turn. Not checked yet New OllamaCloudProvider with dynamic model discovery and automatic context-limit detection. Not checked yet Adds iFlytek Spark and Astron MaaS as supported providers. Not checked yet Adds Fireworks AI as a declarative provider. Not checked yet
+12 more
show less
Adds OpenRouter OpenRouter A unified API gateway that routes requests to dozens of LLM providers under one endpoint, letting cyber tools swap or compare models without rewriting provider-specific integration code. request parameters support. Not checked yet Adds OVHcloud Qwen3.6-27B model with 262K context window context window The maximum amount of text, counted in tokens, a model can take in and produce in one request. Everything the model knows about the task has to fit inside it. . Not checked yet Adds GPT-5.6 family support (gpt-5.6-luna, gpt-5.6-sol, gpt-5.6-terra) with 1,050,000-token context and xhigh reasoning effort. Not checked yet Adds GLM-5.2 model support. Not checked yet Adds TLS support for goose serve ACP sessions. Not checked yet Adds streamable HTTP support in the deeplink generator. Not checked yet Desktop sessions now automatically reconnect after system sleep or connection loss. Not checked yet Chat sessions are now grouped by project in the navigation panel. Not checked yet Delete support for custom apps added to the Apps UI. Not checked yet Desktop localizations added for French, German, Italian, Portuguese, Indonesian, Malay, Vietnamese, and Traditional Chinese. Not checked yet Provider bindings MVP added to goose-sdk with Python wheel publishing. Not checked yet Diagnostics output migrated to structured JSON report format. Not checked yet
17 details in Everything
Goose v1.42.0 adds Ollama Cloud provider, per-message token/cost stats UI, project-grouped sessions, and Python SDK provider bindings.
└──▷ GET THIS VERSION
copy
$ git clone --branch v1.42.0 https://github.com/aaif-goose/goose.git
$ git checkout v1.42.0
Adds OllamaCloudProvider with dynamic model discovery and automatic context-limit detection. Not checked yet Adds per-message usage stats in the desktop UI: token counts, cost, time-to-first-token time-to-first-token Time to first token, the wait between submitting a request and receiving the first generated token. It includes delays such as queuing and processing the prompt. (TTFT TTFT Time to first token, the wait between submitting a request and receiving the first generated token. It includes delays such as queuing and processing the prompt. ), and tokens/sec. Not checked yet Adds per-message usage and cost tracking with derived session totals. Not checked yet Groups chat sessions by project in the navigation panel. Not checked yet Shows project names in the directory-switcher dropdown. Not checked yet
+4 more
show less
Adds provider bindings MVP to goose-sdk with Python wheel publishing. Not checked yet Adds AskAI bot hints for follow-up suggestions. Not checked yet Hardens goose update to fail closed when Sigstore/SLSA SLSA Supply-chain Levels for Software Artifacts, a framework for assessing how software was built and protecting that process from tampering. Provenance records help verify where a build came from. provenance provenance Evidence recording where a software artifact came from and how it was built, including its source and build process. It helps a consumer check that the artifact matches a trusted build. cannot be verified. Not checked yet Increases the /mcp-app-guest body limit to support larger single-file MCP MCP Model Context Protocol, a standard for connecting AI applications to external tools and data sources. A compatible server exposes capabilities an assistant can use through a common interface. apps. Not checked yet
9 details in Everything
Goose v1.41.0 adds TLS for ACP, 15+ new providers, hooks, summon sub-agents, and a wave of new CLI commands.
└──▷ GET THIS VERSION
copy
$ git clone --branch v1.41.0 https://github.com/aaif-goose/goose.git
$ git checkout v1.41.0
└──▷ TRY IT
Serve Goose's ACP endpoint over TLS so remote clients can connect securely.
copy
$ goose serve --tls --tls-cert-path /etc/goose/server.crt --tls-key-path /etc/goose/server.key --port 3284
Edit a previous session's conversation context before forking it, useful for re-running with corrected instructions.
copy
$ goose session --edit --resume <session-id>
Cap tool response payload size to prevent context bloat when tools return large outputs (e.g., grep over a huge codebase).
copy
$ GOOSE_MAX_TOOL_RESPONSE_SIZE=16384 goose run --recipe my-recipe.yaml
Adds --edit session flag to edit a conversation before forking it into a new session. Not checked yet New tui subcommand on goose-cli to launch the terminal UI directly. Not checked yet Supports TLS for goose serve via --tls, --tls-cert-path, and --tls-key-path flags. Not checked yet Adds iFlytek Spark, Astron MaaS, Fireworks AI, Together AI, OrcaRouter, EmpirioLabs AI, Perplexity, Alibaba (Qwen/DashScope), Scaleway, NEAR AI Cloud, Vercel AI Gateway, Databricks AI Gateway, Routstr, FuturMix, oMLX, Atomic Chat, and xAI SuperGrok OAuth as new providers. Not checked yet Adds OpenRouter OpenRouter A unified API gateway that routes requests to dozens of LLM providers under one endpoint, letting cyber tools swap or compare models without rewriting provider-specific integration code. request parameters support for fine-grained API payload control. Not checked yet
+22 more
show less
OpenAI Responses API store param now configurable via OPENAI_STORE env var or config. Not checked yet MLX MLX Apple's machine-learning framework for Apple silicon, which uses the unified memory shared by CPU and GPU to run and fine-tune models on a Mac. model support added for the local inference provider. Not checked yet GLM-5.2 model and gpt-5.5 (ChatGPT Codex) model support added. Not checked yet Summon gains peek mode for async background tasks, context and working_dir parameters for delegate tasks, subagent instructions, and structured metadata for task load results. Not checked yet Session import functionality added. Not checked yet OTLP OTLP OpenTelemetry Protocol, a standard way to send traces, metrics and logs between applications, collectors and monitoring systems. It lets telemetry move across compatible tools. logging schema added for cross-tool detection. Not checked yet Worktree-aware directory switcher added. Not checked yet Chat history search feature added. Not checked yet TUI TUI A text-based user interface that runs inside a terminal, using characters and ANSI escape codes to render menus, panes, and widgets without a graphical desktop, letting cyber tools ship rich interactivity with no GUI dependency. diff viewer added. Not checked yet Paginated session list in ACP. Not checked yet ACP schema now exposes available tools. Not checked yet ACP thinking effort config option added. Not checked yet Structured per-provider config block with non-destructive provider switching. Not checked yet Language selection feature in desktop app with locales for French, German, Italian, Portuguese, Indonesian, Malay, Vietnamese, Traditional Chinese, Spanish, Korean, Japanese, Hindi, Russian, Turkish, and Simplified Chinese. Not checked yet Stable OAuth redirect URI via GOOSE_OAUTH_CALLBACK_PORT environment variable. Not checked yet Unified thinking effort control across all providers. Not checked yet Image read tool added. Not checked yet Open-plugins generalization plus skills support. Not checked yet Option to disable automatic update downloads. Not checked yet Last message snippets for ACP sessions. Not checked yet GOOSE_FAST_MODEL environment variable support for specifying a fast model. Not checked yet Optional api_key configuration for declarative OpenAI-engine providers. Not checked yet
27 details in Everything
Goose v1.40.0 adds 13+ new AI providers, hooks system with tool-use denial, local code review, new CLI slash commands, and prompt injection prompt injection An attack where text the model reads, such as a web page, file or tool result, carries instructions that override the user's. It is the main security risk for agents with tools. mitigation.
└──▷ GET THIS VERSION
copy
$ git clone --branch v1.40.0 https://github.com/aaif-goose/goose.git
$ git checkout v1.40.0
└──▷ TRY IT
Cap how much output a noisy tool can return so long responses don't blow the context window context window The maximum amount of text, counted in tokens, a model can take in and produce in one request. Everything the model knows about the task has to fit inside it. .
copy
$ GOOSE_MAX_TOOL_RESPONSE_SIZE=32768 goose run --recipe audit.yaml
Run an AI-powered local code review on the current working directory before opening a pull request.
New /status slash command in the CLI to check session state at a glance. Not checked yet New goose review subcommand for running local AI-powered code review. Not checked yet Adds a hooks system for extensibility, including PreToolUse denial to block tool calls before execution and blocking Stop hook decisions. Not checked yet Prompt injection mitigation enabled by default. Not checked yet Supports 13+ new model providers: Together AI, EmpirioLabs, OrcaRouter, NEAR AI Cloud, Perplexity, Alibaba/Qwen via DashScope, Databricks AI Gateway, Scaleway, Vercel AI Gateway, xAI SuperGrok OAuth, OpenAI GPT via AWS Bedrock mantle endpoint, GLM-5.2, and MLX MLX Apple's machine-learning framework for Apple silicon, which uses the unified memory shared by CPU and GPU to run and fine-tune models on a Mac. models for local inference. Not checked yet
+11 more
show less
New GOOSE_FAST_MODEL environment variable to designate a fast model for lightweight operations. Not checked yet Unified thinking effort control and canonical thinking modes across all supported providers. Not checked yet Cache token tracking for accurate per-session cost reporting. Not checked yet Encrypted Nostr session sharing. Not checked yet Peek mode for async background tasks in summon, plus context parameter and working_dir override for delegate tasks. Not checked yet New TUI TUI A text-based user interface that runs inside a terminal, using characters and ANSI escape codes to render menus, panes, and widgets without a graphical desktop, letting cyber tools ship rich interactivity with no GUI dependency. subcommand (goose-cli tui) with built-in diff viewer. Not checked yet Chat history search in the desktop navigation panel. Not checked yet Session import and deep links (goose://new-session, goose://resume) including initial prompt support. Not checked yet Image read tool for ingesting local images. Not checked yet Streamable HTTP Streamable HTTP An MCP transport protocol that lets a client send JSON-RPC requests over HTTP POST while the server streams responses back using Server-Sent Events, replacing the older stdio and SSE transports for remote tool communication. support in the deeplink generator. Not checked yet Linux musl CLI builds for broader Linux compatibility. Not checked yet
16 details in Everything
Goose v1.39.0 adds /status and /model CLI commands, MLX MLX Apple's machine-learning framework for Apple silicon, which uses the unified memory shared by CPU and GPU to run and fine-tune models on a Mac. local inference, 8 new providers, Azure Entra ID Entra ID Microsoft's cloud-based identity and access management service, formerly Azure Active Directory, used by cyber tools to authenticate users and enforce access policies in Microsoft 365 and Azure environments. auth, and a unified OTLP OTLP OpenTelemetry Protocol, a standard way to send traces, metrics and logs between applications, collectors and monitoring systems. It lets telemetry move across compatible tools. logging schema.
└──▷ GET THIS VERSION
copy
$ git clone --branch v1.39.0 https://github.com/aaif-goose/goose.git
$ git checkout v1.39.0
└──▷ TRY IT
Check session status interactively without leaving a running CLI session.
Switch the active model mid-session to compare responses without restarting.
Run a recipe and print token/generation statistics on completion to benchmark cost and speed.
copy
$ goose run --recipe goose-self-test.yaml --stats
Adds /status slash command to the CLI for quick session status reporting. Not checked yet Adds /model slash command to the CLI for switching models mid-session without restarting. Not checked yet Adds --stats flag to print generation statistics after a run completes. Not checked yet Supports MLX models for local inference, enabling Apple Silicon on-device model execution. Not checked yet Supports OpenAI GPT models via AWS Bedrock mantle endpoint. Not checked yet
+23 more
show less
Adds Azure Entra ID bearer token bearer token A credential that grants access to whoever presents it, without a separate proof of possession. Protecting the token matters because a stolen copy can be used by someone else. authentication via AZURE_OPENAI_AD_TOKEN environment variable. Not checked yet Adds EmpirioLabs, OrcaRouter, Together AI, Perplexity, Alibaba (Qwen/DashScope), Databricks AI Gateway, and xAI SuperGrok as new declarative providers. Not checked yet Adds HuggingFace OAuth support with a dedicated auth tab in settings. Not checked yet Introduces GOOSE_SERVER__SECRET_KEY support at the goose serve ACP endpoint for API authentication. Not checked yet Adds configurable tool output size limit via GOOSE_MAX_TOOL_RESPONSE_SIZE environment variable. Not checked yet Adds configurable per-agent timeout for Open Model Gym via GYM_AGENT_TIMEOUT environment variable. Not checked yet Loads global agent hints automatically from ~/.agents/AGENTS.md. Not checked yet Adds an image read tool for processing image files directly in sessions. Not checked yet Adds peek mode for async background tasks in summon, enabling non-blocking status checks. Not checked yet Adds context and working_dir override parameters for summon delegate tasks. Not checked yet Adds structured metadata to summon task load results for richer programmatic consumption. Not checked yet Adds a quarterly scheduling option to the task scheduler. Not checked yet Adds a unified OTLP logging schema for cross-tool detection and observability. Not checked yet Enables canonical thinking modes configuration and ACP thinking effort config option. Not checked yet Enables prompt injection prompt injection An attack where text the model reads, such as a web page, file or tool result, carries instructions that override the user's. It is the main security risk for agents with tools. mitigation by default for internal users. Not checked yet Adds _meta.conversationBefore for ACP session forking. Not checked yet Adds ACP session search, list with keyword/type filter, session import, and system prompt setter in the Desktop UI. Not checked yet Adds MCP MCP Model Context Protocol, a standard for connecting AI applications to external tools and data sources. A compatible server exposes capabilities an assistant can use through a common interface. extensions support in open plugins. Not checked yet Enables goose://new-session deep link for pre-populated session launch. Not checked yet Adds Korean, Japanese, Hindi, Spanish, Russian, and Turkish localization support. Not checked yet Adds CLI command to list skills with token counts. Not checked yet Surfaces Anthropic stream refusals as visible errors instead of silent failures. Not checked yet Tracks cache tokens for accurate cost reporting. Not checked yet
28 details in Everything
Goose v1.38.0 adds goose review, hooks, 8 new providers, ACP secret-key auth, egress directionality, and a PreToolUse denial hook.
└──▷ GET THIS VERSION
copy
$ git clone --branch v1.38.0 https://github.com/aaif-goose/goose.git
$ git checkout v1.38.0
└──▷ TRY IT
Run an AI code review on the current repo, capping each check subprocess to 10 turns to prevent runaway agents.
copy
$ goose review --turn-limit 10
Start the ACP server with a secret key so only authenticated clients can connect.
copy
$ GOOSE_SERVER__SECRET_KEY=mysecretkey goose serve
Prevent context bloat from large tool responses by capping output size, and disable noisy per-call summary cards.
copy
$ GOOSE_MAX_TOOL_RESPONSE_SIZE=8192 GOOSE_DISABLE_TOOL_CALL_SUMMARY=true goose run --recipe my-recipe.yaml
Adds goose review command for local AI-powered code review with per-check turn limits and orchestrated subprocesses. Not checked yet New hooks system with PreToolUse denial hook to intercept and block tool calls before execution. Not checked yet Adds GOOSE_SERVER__SECRET_KEY support at the goose serve ACP endpoint for authenticated API access. Not checked yet Adds egress logging directionality (outbound/inbound/unknown) to security tracing events, enabling data-exfiltration vs. data-fetch distinction. Not checked yet Unified OTLP OTLP OpenTelemetry Protocol, a standard way to send traces, metrics and logs between applications, collectors and monitoring systems. It lets telemetry move across compatible tools. logging schema for cross-tool detection across security monitoring pipelines. Not checked yet
+27 more
show less
New /model slash command in CLI sessions for switching models mid-conversation without restarting. Not checked yet New /goal slash command for agent self-evaluation before finishing a task. Not checked yet Adds GOOSE_MAX_TOOL_RESPONSE_SIZE env var to cap tool output size and prevent context bloat. Not checked yet Adds GOOSE_DISABLE_TOOL_CALL_SUMMARY env var to opt out of per-tool-call summary cards. Not checked yet Adds GOOSE_OAUTH_CALLBACK_PORT env var for a stable OAuth redirect URI across sessions. Not checked yet New image read tool enabling agents to process and reason over image files. Not checked yet Session import now supports Claude Code, Codex, and Pi .jsonl.jsonl A text format where each line is a self-contained, valid JSON object, making it easy to stream, append, and process large datasets line by line without parsing a full JSON array into memory. transcript formats in addition to Goose exports. Not checked yet Encrypted Nostr session sharing for secure cross-device session transfer. Not checked yet Supports 8 new providers: Perplexity, Alibaba/Qwen via DashScope, Databricks AI Gateway, NEAR AI Cloud, Scaleway, Routstr, Vercel AI Gateway, and xAI SuperGrok OAuth. Not checked yet Adds Hugging Face OAuth support with a new Settings > Auth tab for credential management. Not checked yet Canonical thinking modes available across all supported providers. Not checked yet Steering messages via ACP enable mid-session behavioral guidance from ACP clients. Not checked yet ACP session management (list, search, filter by keyword/type, paginate, get session info) now available in Desktop and via API. Not checked yet New Scholar Sidekick MCP MCP Model Context Protocol, a standard for connecting AI applications to external tools and data sources. A compatible server exposes capabilities an assistant can use through a common interface. extension for research workflows. Not checked yet Linux musl CLI builds for glibc-free deployments. Not checked yet Windows CUDA CUDA NVIDIA's platform and programming interface for running general computation on its GPUs. Nearly every AI training and inference stack is built on it first. release artifacts for GPU GPU A graphics processing unit, a processor that performs many calculations in parallel. AI workloads use GPUs to speed up model training and inference. -accelerated local inference. Not checked yet Nushell terminal and tab-completion support. Not checked yet Quarterly scheduling option for recurring recipes. Not checked yet New goose://resume and goose://new-session deep links for Desktop session navigation. Not checked yet Projects as backend sources with system prompt injection prompt injection An attack where text the model reads, such as a web page, file or tool result, carries instructions that override the user's. It is the main security risk for agents with tools. for scoped agent contexts. Not checked yet Opt-in GOOSE_FAST_MODEL env var honored in ModelConfig::with_fast for fast-model overrides. Not checked yet Harbor eval runner for structured agent benchmarking. Not checked yet CLI command to list skills with token counts for context budget planning. Not checked yet TUI TUI A text-based user interface that runs inside a terminal, using characters and ANSI escape codes to render menus, panes, and widgets without a graphical desktop, letting cyber tools ship rich interactivity with no GUI dependency. command (goose tui) and TUI diff viewer added to goose-cli. Not checked yet Configurable MAX_CODE_BLOCK_LINES via env var to control output verbosity. Not checked yet Minimal UniFFI setup enabling cross-language SDK bindings. Not checked yet Simplified Chinese (zh-CN) UI translation. Not checked yet
32 details in Everything
Goose v1.37.0 adds 9+ new providers, PreToolUse hooks, goose review, /model switching, skills listing, and encrypted session sharing.
└──▷ GET THIS VERSION
copy
$ git clone --branch v1.37.0 https://github.com/aaif-goose/goose.git
$ git checkout v1.37.0
└──▷ TRY IT
Run an interactive code review on your current working directory to get AI feedback before opening a PR.
Switch to a different model mid-session without losing context — useful when a task needs a more capable or faster model.
copy
$ /model claude-opus-4-5
Audit token cost of installed skills to understand context budget usage before starting a long session.
Cap how much output a single tool call can return to prevent context overflow from verbose tools.
copy
$ GOOSE_MAX_TOOL_RESPONSE_SIZE=50000 goose run --recipe my-recipe.yaml
Adds goose review command for local code review directly from the CLI. Not checked yet Adds /model slash command to switch the active model mid-session without restarting. Not checked yet Adds goose skills list subcommand to display all installed skills with token counts. Not checked yet Adds PreToolUse denial hooks so practitioners can intercept and block tool calls before execution. Not checked yet Adds a Hooks system for event-driven automation around agent actions. Not checked yet
+32 more
show less
Adds /goal slash command for agent self-evaluation before finishing a task. Not checked yet Adds configurable tool output size limit via GOOSE_MAX_TOOL_RESPONSE_SIZE environment variable. Not checked yet Adds encrypted Nostr session sharing for secure cross-device session handoff. Not checked yet Adds xAI SuperGrok OAuth subscription provider. Not checked yet Adds Databricks AI Gateway (databricks_v2) provider with routing across OpenAI Responses, Anthropic Messages, and MLflow chat completions. Not checked yet Adds Perplexity as a declarative OpenAI-compatible provider. Not checked yet Adds Alibaba (Qwen via DashScope) declarative provider. Not checked yet Adds NEAR AI Cloud, Scaleway, Routstr, FuturMix, oMLX, Atomic Chat, and Vercel AI Gateway as new declarative providers. Not checked yet Adds Scholar Sidekick MCP MCP Model Context Protocol, a standard for connecting AI applications to external tools and data sources. A compatible server exposes capabilities an assistant can use through a common interface. extension. Not checked yet Adds GitHub Copilot /responses API support. Not checked yet Adds Linux Vulkan support for local inference. Not checked yet Adds goose://resume and goose://new-session deep links for desktop session management. Not checked yet Adds quarterly scheduling option for the recipe scheduler. Not checked yet Adds optional --parameters flag to scheduled recipes. Not checked yet Adds ACP session system prompt setter for programmatic prompt injection prompt injection An attack where text the model reads, such as a web page, file or tool result, carries instructions that override the user's. It is the main security risk for agents with tools. . Not checked yet Adds ACP session list pagination. Not checked yet Exposes raw provider supported models over ACP. Not checked yet Adds GOOSE_FAST_MODEL env var to configure the fast model in ModelConfig::with_fast. Not checked yet Adds configurable MAX_CODE_BLOCK_LINES via environment variable. Not checked yet Adds GOOSE_OAUTH_CALLBACK_PORT env var for a stable OAuth redirect URI. Not checked yet Adds proactive OAuth token refresh to avoid re-authentication on every session. Not checked yet Adds TUI TUI A text-based user interface that runs inside a terminal, using characters and ANSI escape codes to render menus, panes, and widgets without a graphical desktop, letting cyber tools ship rich interactivity with no GUI dependency. command (goose tui) gated behind a tui feature flag. Not checked yet Adds TUI diff viewer for reviewing file changes in the terminal UI. Not checked yet Adds Nushell terminal and shell completion support. Not checked yet Adds Russian language and Turkish desktop locale support. Not checked yet Adds Simplified Chinese (zh-CN) translation. Not checked yet Adds Harbor eval runner for benchmarking. Not checked yet Adds summon subagent instructions for multi-agent orchestration. Not checked yet Adds slash commands (built-in, skill, recipe) in the ACP server. Not checked yet Adds projects as backend sources with system prompt injection. Not checked yet Groups consecutive tool calls into one summarized chain card in the UI. Not checked yet Strips chain-of-thought chain-of-thought Intermediate reasoning a model writes out before its final answer. Reasoning models generate it automatically and it is often billed as extra output tokens. markers from custom provider output automatically. Not checked yet
37 details in Everything
Goose v1.36.0 adds goose review, a hooks system for PreToolUse denial, TUI TUI A text-based user interface that runs inside a terminal, using characters and ANSI escape codes to render menus, panes, and widgets without a graphical desktop, letting cyber tools ship rich interactivity with no GUI dependency. diff viewer, /goal self-evaluation, and 7+ new AI providers.
└──▷ GET THIS VERSION
copy
$ git clone --branch v1.36.0 https://github.com/block/goose.git
$ git checkout v1.36.0
└──▷ TRY IT
Run a local code review on your current working directory before opening a PR.
Suppress verbose tool call summaries in CI pipelines where only final output matters.
copy
$ GOOSE_DISABLE_TOOL_CALL_SUMMARY=1 goose run --recipe my-recipe.yaml
New goose review command for local code review without leaving the terminal. Not checked yet New TUI diff viewer to inspect file changes interactively in the terminal UI. Not checked yet New /goal slash command lets the agent self-evaluate against a stated goal before finishing a session. Not checked yet New hooks system supports PreToolUse denial and extensibility, enabling policy enforcement on tool calls. Not checked yet Adds NEAR AI Cloud, Scaleway, Vercel AI Gateway, Atomic Chat, Routstr, FuturMix, and oMLX as new AI providers. Not checked yet
+23 more
show less
Slash commands (built-in, skill, recipe) now available in the ACP server. Not checked yet Summon subagent instructions enable structured multi-agent delegation. Not checked yet Unified thinking effort control across all providers via a single configuration surface. Not checked yet New goose://resume and goose://new-session deep links for session management from external apps. Not checked yet New --parameters flag for scheduled recipes to pass runtime parameters. Not checked yet Quarterly scheduling option added to the recipe scheduler. Not checked yet GitHub Copilot /responses API support added. Not checked yet Chat history search feature added to the desktop UI. Not checked yet Projects can now act as backend sources with system prompt injection prompt injection An attack where text the model reads, such as a web page, file or tool result, carries instructions that override the user's. It is the main security risk for agents with tools. . Not checked yet Linux Vulkan support enables GPU GPU A graphics processing unit, a processor that performs many calculations in parallel. AI workloads use GPUs to speed up model training and inference. -accelerated local inference on Linux. Not checked yet Linux musl CLI builds available for Alpine/musl-based environments. Not checked yet Windows CUDA CUDA NVIDIA's platform and programming interface for running general computation on its GPUs. Nearly every AI training and inference stack is built on it first. release artifacts now published for GPU-accelerated local inference on Windows. Not checked yet New GOOSE_DISABLE_TOOL_CALL_SUMMARY env var to suppress tool call summaries. Not checked yet New GOOSE_OAUTH_CALLBACK_PORT env var for a stable OAuth redirect URI across sessions. Not checked yet Proactive OAuth token refresh avoids re-authentication on every new session. Not checked yet New MAX_CODE_BLOCK_LINES env var controls code block rendering length. Not checked yet Honors GOOSE_FAST_MODEL env var in ModelConfig for fast-model selection. Not checked yet GPT-5.5 added to known models list. Not checked yet ACP session list now supports pagination. Not checked yet Encrypted Nostr session sharing for secure cross-client session handoff. Not checked yet Nushell terminal and shell completion support added. Not checked yet Worktree-aware directory switcher for multi-worktree Git workflows. Not checked yet Harbor eval runner integration added. Not checked yet
28 details in Everything
Goose v1.35.0 adds a hooks system for pre/post tool execution control, goose review, /goal self-evaluation, and five new providers.
└──▷ GET THIS VERSION
copy
$ git clone --branch v1.35.0 https://github.com/block/goose.git
$ git checkout v1.35.0
└──▷ TRY IT
Run a local code review on your current working directory to catch issues before committing.
Adds a hooks system for extensible pre/post tool execution, including PreToolUse denial hooks to block tool calls before they run. Not checked yet New goose review command for local code review directly from the CLI. Not checked yet New /goal slash command lets the agent self-evaluate against its goal before finishing a session. Not checked yet Diff viewer added to the TUI TUI A text-based user interface that runs inside a terminal, using characters and ANSI escape codes to render menus, panes, and widgets without a graphical desktop, letting cyber tools ship rich interactivity with no GUI dependency. for inspecting code changes inline. Not checked yet Unified thinking effort control across all providers for consistent reasoning depth. Not checked yet
+27 more
show less
Adds GOOSE_DISABLE_TOOL_CALL_SUMMARY environment variable to opt out of per-tool-call summaries. Not checked yet Groups consecutive tool calls into a single summarized chain card in the UI. Not checked yet Supports --parameters flag for scheduled recipes. Not checked yet Adds quarterly scheduling option for recipes. Not checked yet New goose://new-session deep link opens a fresh chat window. Not checked yet Adds GOOSE_OAUTH_CALLBACK_PORT environment variable for a stable OAuth redirect URI. Not checked yet Chat history search added to the navigation panel. Not checked yet Encrypted Nostr session sharing for secure collaboration. Not checked yet Nushell terminal and shell completion support. Not checked yet Summon subagent instructions for directing spawned subagents. Not checked yet Open-plugins generalization with skills support. Not checked yet Projects usable as backend sources with system prompt injection prompt injection An attack where text the model reads, such as a web page, file or tool result, carries instructions that override the user's. It is the main security risk for agents with tools. . Not checked yet Worktree-aware directory switcher in the TUI. Not checked yet Slash commands (built-in, skill, recipe) now available in the ACP server. Not checked yet Paginated session list in ACP. Not checked yet Recipe discovery and execution in the ACP server. Not checked yet Agents CRUD CRUD The four basic database operations — Create, Read, Update, Delete — used as shorthand for any system that manages persistent data records, giving cyber tools a standard model for interacting with stored information. operations via ACP. Not checked yet ACP streamable HTTP spec compliance. Not checked yet New Vercel AI Gateway provider with opt-in leaderboard participation. Not checked yet New GitHub Copilot /responses API support. Not checked yet GPT-5.5 added to known models. Not checked yet Linux musl CLI builds now available. Not checked yet Linux Vulkan support for local inference. Not checked yet Windows CUDA CUDA NVIDIA's platform and programming interface for running general computation on its GPUs. Nearly every AI training and inference stack is built on it first. release artifacts. Not checked yet New providers: Atomic Chat, Routstr, FuturMix, oMLX, SaladCloud AI Gateway. Not checked yet Simplified Chinese (zh-CN) translation added. Not checked yet Strip chain-of-thought chain-of-thought Intermediate reasoning a model writes out before its final answer. Reasoning models generate it automatically and it is often billed as extra output tokens. markers from custom provider output. Not checked yet
32 details in Everything
Goose v1.34.0 adds hooks, agent CRUD CRUD The four basic database operations — Create, Read, Update, Delete — used as shorthand for any system that manages persistent data records, giving cyber tools a standard model for interacting with stored information. , project-based system prompts, a TUI TUI A text-based user interface that runs inside a terminal, using characters and ANSI escape codes to render menus, panes, and widgets without a graphical desktop, letting cyber tools ship rich interactivity with no GUI dependency. binary, and Linux Vulkan local inference.
└──▷ GET THIS VERSION
copy
$ git clone --branch v1.34.0 https://github.com/block/goose.git
$ git checkout v1.34.0
Adds hooks support for customizable agent behavior at runtime. Not checked yet Adds CRUD operations for managing agents via ACP. Not checked yet Enables projects as backend sources with system prompt injection prompt injection An attack where text the model reads, such as a web page, file or tool result, carries instructions that override the user's. It is the main security risk for agents with tools. into sessions. Not checked yet Supports auto-updating for plugins installed to ~/.agents/plugins. Not checked yet Adds Linux Vulkan support for local inference workloads. Not checked yet
+12 more
show less
Achieves ACP streamable HTTP spec compliance. Not checked yet Exposes built-in skills through the ACP sources list. Not checked yet Introduces a provider-first onboarding flow for new users. Not checked yet Introduces configurable Ollama host in goose2. Not checked yet Adds Google model inventory refresh support. Not checked yet Adds skills picker directly in the chat composer. Not checked yet Supports mergeable configs for layered configuration management. Not checked yet Adds MCP MCP Model Context Protocol, a standard for connecting AI applications to external tools and data sources. A compatible server exposes capabilities an assistant can use through a common interface. Apps payload hydration and replay in goose2. Not checked yet Translates ACP host capabilities into MCP initialization for MCP Apps. Not checked yet Releases Windows CUDA CUDA NVIDIA's platform and programming interface for running general computation on its GPUs. Nearly every AI training and inference stack is built on it first. build artifacts for local inference on Windows. Not checked yet Introduces the goose-tui binary as a standalone terminal UI. Not checked yet Groups consecutive tool calls into a single summarized chain card in the UI. Not checked yet
17 details in Everything
Goose v1.32.0 adds Exa AI search, two new AI providers, @agent mentions, /skills and /edit commands, voice dictation, and Unix socket MCP MCP Model Context Protocol, a standard for connecting AI applications to external tools and data sources. A compatible server exposes capabilities an assistant can use through a common interface. transport.
└──▷ GET THIS VERSION
copy
$ git clone --branch v1.32.0 https://github.com/block/goose.git
$ git checkout v1.32.0
Adds Exa AI-powered web search as a built-in tool. Not checked yet New /skills command to browse and manage available agent skills from the CLI. Not checked yet New /edit command for on-demand prompt editing mid-session. Not checked yet Supports @agent mentions in chat to target specific agents directly. Not checked yet Adds Novita AI as a declarative provider. Not checked yet
+12 more
show less
Adds Kimi Code as a provider with OAuth device-flow OAuth device-flow An OAuth 2.0 grant type that lets a device with no browser obtain an access token by displaying a short code the user enters on a separate device, enabling auth on CLIs and headless systems. authentication. Not checked yet Adds Cash App extension to the Extensions Library. Not checked yet Adds Unix domain socket transport support for StreamableHttp MCP connections. Not checked yet Client ID Metadata Document Client ID Metadata Document An OAuth 2.0 mechanism letting a client publish its own metadata (redirect URIs, grant types, contacts) at a well-known URL so authorization servers can fetch and verify it without manual registration. (CIMD CIMD A text-based Short Message Service protocol from Nokia for exchanging SMS messages between applications and SMS centers, used to build tools that send, receive, or test carrier messaging flows. ) support for OAuth flows. Not checked yet Desktop notifications when Goose finishes a task. Not checked yet Auto-compaction for the context window context window The maximum amount of text, counted in tokens, a model can take in and produce in one request. Everything the model knows about the task has to fit inside it. to manage long sessions automatically. Not checked yet Context window usage now shown inline in the status bar and chat input. Not checked yet Voice dictation support via direct-ACP pattern in goose2. Not checked yet Chat attachments support in goose2. Not checked yet Extension management screen added to the TUI TUI A text-based user interface that runs inside a terminal, using characters and ANSI escape codes to render menus, panes, and widgets without a graphical desktop, letting cyber tools ship rich interactivity with no GUI dependency. . Not checked yet Extensions settings page and context panel widget added. Not checked yet Threads can now be associated with projects. Not checked yet
17 details in Everything
Goose v1.30.0 adds goose serve, goose doctor, egress logging, new Gemini/Copilot/Zhipu providers, and major TUI TUI A text-based user interface that runs inside a terminal, using characters and ANSI escape codes to render menus, panes, and widgets without a graphical desktop, letting cyber tools ship rich interactivity with no GUI dependency. improvements.
└──▷ GET THIS VERSION
copy
$ git clone --branch v1.30.0 https://github.com/block/goose.git
$ git checkout v1.30.0
└──▷ TRY IT
Run Goose as a persistent background service, useful for integrations or automation pipelines.
Diagnose misconfigured environments or credential issues before starting a session.
Prevent tool output from being truncated when inspecting verbose command results during a session.
copy
$ GOOSE_SHOW_FULL_OUTPUT=true goose run --text 'List all open ports and their processes'
New goose serve subcommand runs Goose as a background service. Not checked yet New goose doctor command diagnoses common configuration and environment issues. Not checked yet New egress logging inspector logs and inspects all outbound network calls for security visibility. Not checked yet Adds Gemini OAuth provider for first-class OAuth-based authentication with Google Gemini. Not checked yet Adds Copilot ACP provider for GitHub Copilot via ACP. Not checked yet
+23 more
show less
Adds Zhipu AI as a declarative provider with configurable ZHIPU_BASE_URL override. Not checked yet Adds Gemma 4 as a local model option. Not checked yet Supports configurable fast_model for declarative providers to designate a lightweight model for quick tasks. Not checked yet Tab-expandable tool call output collapses and expands cleanly in the TUI. Not checked yet Independent --text mode decouples headless text output from the normal TUI. Not checked yet New GOOSE_SHOW_FULL_OUTPUT config option disables tool output truncation. Not checked yet GOOSE_CLI_SHOW_THINKING can now be set in config.yaml in addition to env vars. Not checked yet GOOSE_CONTEXT_LIMIT is now readable from config.yaml. Not checked yet GOOSE_DISABLE_KEYRING is now honoured from config.yaml at startup. Not checked yet Configurable per-extension timeouts via ACP _meta or a global default. Not checked yet Skills are now visible and browsable from the desktop app UI. Not checked yet Skills are now managed as a dedicated platform extension. Not checked yet Recursive skill discovery finds skills at any depth under configured roots. Not checked yet Adds react-intl-based internationalization infrastructure to the desktop app. Not checked yet Per-delegate max_turns override is now supported. Not checked yet skip_canonical_filtering option added for declarative providers. Not checked yet Optional native-tls support added as an alternative to rustls for system-TLS environments. Not checked yet AWS provider dependencies are now feature-gated behind the aws-providers feature flag for leaner builds. Not checked yet Telemetry is now behind Cargo feature flags. Not checked yet Parallelized extension loading in ACP server for faster startup. Not checked yet Secrets files are now created with tighter OS-level permissions. Not checked yet Adds Rust ACP client for extension methods. Not checked yet Adds Inference Mesh settings tab. Not checked yet
28 details in Everything
Goose v1.29.0 adds orchestration support, Gemini-ACP provider, GOOSE_SHELL config, Sigstore provenance provenance Evidence recording where a software artifact came from and how it was built, including its source and build process. It helps a consumer check that the artifact matches a trusted build. verification, and stabilized ACP session APIs.
└──▷ GET THIS VERSION
copy
$ git clone --branch v1.29.0 https://github.com/block/goose.git
$ git checkout v1.29.0
└──▷ TRY IT
Override the shell Goose uses for command execution — useful when your default shell lacks the environment your tasks need.
copy
$ GOOSE_SHELL=/usr/bin/zsh goose run --session my-session
Adds GOOSE_SHELL environment variable to configure the preferred shell used by Goose. Not checked yet Adds Sigstore/SLSA SLSA Supply-chain Levels for Software Artifacts, a framework for assessing how software was built and protecting that process from tampering. Provenance records help verify where a build came from. provenance verification for Rust self-update, strengthening supply-chain integrity. Not checked yet Adds a Gemini-ACP provider with subscription model support. Not checked yet Adds orchestration support for coordinating multi-agent workflows. Not checked yet Stabilizes ACP session/set_config, list, delete, and close endpoints; adds set_config to the ACP session API. Not checked yet
+8 more
show less
Adds ability to manage sub-recipes in the desktop UI. Not checked yet Displays delegate sub-agent logs in the desktop UI. Not checked yet Adds expandable sidebar to show full chat names in the desktop UI. Not checked yet Adds tool filtering by MCP MCP Model Context Protocol, a standard for connecting AI applications to external tools and data sources. A compatible server exposes capabilities an assistant can use through a common interface. Apps visibility metadata. Not checked yet Passes toolInfo to MCP Apps via hostContext. Not checked yet Adds CodeMode ToolDisclosure variants support. Not checked yet Feature-gates local inference dependencies for leaner builds. Not checked yet Upgrades ink renderer for performance and switches CLI from HTTP to stdio transport. Not checked yet
13 details in Everything
Goose v1.28.0 adds adversarial safety agent, Claude adaptive thinking, MCP MCP Model Context Protocol, a standard for connecting AI applications to external tools and data sources. A compatible server exposes capabilities an assistant can use through a common interface. Roots, new LLM LLM A large language model, a neural network trained on large amounts of text to recognize patterns and generate language. Applications use it for tasks such as answering, summarizing and writing code. providers, and ACP filesystem delegation.
└──▷ GET THIS VERSION
copy
$ git clone --branch v1.28.0 https://github.com/block/goose.git
$ git checkout v1.28.0
Adds an adversarial agent that monitors sessions to prevent information leakage and other unsafe behaviors. Not checked yet Supports Claude adaptive thinking, enabling dynamic reasoning depth for Claude models. Not checked yet Introduces a new onboarding flow for first-time setup. Not checked yet Persists GooseMode per-session via the session database so mode selection survives restarts. Not checked yet Propagates session.id to OpenTelemetry OpenTelemetry An open standard and set of tools for collecting traces, metrics and logs from applications. It lets services export observability data without tying instrumentation to one monitoring vendor. spans and log records for end-to-end tracing. Not checked yet
+16 more
show less
Migrates the autovisualiser extension to MCP Apps. Not checked yet Adds ACP session/set_mode handler for programmatic mode switching via ACP. Not checked yet Supports MCP Roots, enabling scoped filesystem access declarations for MCP servers. Not checked yet Loads hint files from nested subdirectories, not just the top-level directory. Not checked yet Adds ACP read tool for delegating filesystem I/O to ACP clients. Not checked yet Adds a summarize tool for deterministic, reproducible file reads. Not checked yet Adds ACP providers for Claude Code and Codex (claude-acp, codex-acp). Not checked yet Captures reasoning summaries from the OpenAI Responses API. Not checked yet Adds MiniMax as a provider via an Anthropic-compatible API. Not checked yet Adds Tensorix as a declarative LLM provider. Not checked yet Supports individually loadable skill files via load() in summon. Not checked yet Adds Avian as an LLM provider. Not checked yet Adds a base path field for custom provider configuration. Not checked yet Improves private channel detection for the Slack bot integration. Not checked yet Enables preferred microphone selection for voice input. Not checked yet Supports secret interpolation in streamable HTTP extension URLs. Not checked yet
21 details in Everything
Goose v1.27.0 adds structured shell output, tree-sitter tree-sitter A parser-generator library, maintained by the Tree-sitter project, that builds fast, incremental, error-tolerant syntax trees for source code; cyber tools use it to parse and query code without writing a full compiler frontend. AST AST An abstract syntax tree, a structured representation of parsed code. Tools inspect or transform this tree to understand expressions and statements rather than just matching text. analysis, Anthropic adaptive thinking, and self-signed HTTPS HTTPS HTTP carried over an encrypted TLS connection. It protects traffic in transit and lets a browser verify the server's certificate. for goosed.
└──▷ GET THIS VERSION
copy
$ git clone --branch v1.27.0 https://github.com/block/goose.git
$ git checkout v1.27.0
└──▷ TRY IT
Run a shell command and consume the structured stdout/stderr output separately for automated analysis or CI pipelines.
copy
$ # When Goose executes shell commands, the tool now returns:
# { "stdout": "...", "stderr": "..." }
# Reference stdout or stderr independently in follow-up instructions or recipes.
Adds structured {stdout, stderr} return from the shell tool with an explicit output schema, making it easier to parse command results programmatically. Not checked yet New platform extension with tree-sitter AST parsing enables deeper code analysis within Goose sessions. Not checked yet Supports Anthropic adaptive (extended) thinking mode for compatible Claude models. Not checked yet Adds fullscreen and Picture-in-Picture (PiP) display modes for MCP MCP Model Context Protocol, a standard for connecting AI applications to external tools and data sources. A compatible server exposes capabilities an assistant can use through a common interface. Apps. Not checked yet Introduces self-signed HTTPS support for the goosed server, enabling secure local deployments. Not checked yet
+3 more
show less
Adds code-mode feature flag to include pctx/Code Mode as an optional dependency. Not checked yet Adds base_path field to custom provider config for more flexible provider routing. Not checked yet Adds GPT 5-3-Codex model support in the Databricks provider. Not checked yet
8 details in Everything
Goose v1.26.0 adds local llama.cpp inference, Cerebras/Moonshot/LMStudio providers, Telegram gateway, and sub-agent model config.
└──▷ GET THIS VERSION
copy
$ git clone --branch v1.26.0 https://github.com/block/goose.git
$ git checkout v1.26.0
└──▷ TRY IT
Pin sub-agent tasks to a cheaper or faster model while keeping your primary agent on a more capable one.
copy
$ GOOSE_SUBAGENT_MODEL=claude-haiku-4-5 GOOSE_SUBAGENT_PROVIDER=anthropic goose session start
Adds local inference provider backed by llama.cpp with HuggingFace model management. Not checked yet Adds Cerebras provider support. Not checked yet Adds Moonshot and Kimi Code as declarative providers. Not checked yet Adds LMStudio as a declarative provider. Not checked yet Adds a gateway to chat with Goose via Telegram and similar messaging platforms. Not checked yet
+17 more
show less
Adds MCP MCP Model Context Protocol, a standard for connecting AI applications to external tools and data sources. A compatible server exposes capabilities an assistant can use through a common interface. apps sampling support. Not checked yet Adds Neighborhood extension to the Extensions Library. Not checked yet Overhauled computer controller with Peekaboo integration. Not checked yet Adds a TUI TUI A text-based user interface that runs inside a terminal, using characters and ANSI escape codes to render menus, panes, and widgets without a graphical desktop, letting cyber tools ship rich interactivity with no GUI dependency. client for goose-acp. Not checked yet Adds GOOSE_SUBAGENT_MODEL and GOOSE_SUBAGENT_PROVIDER config options to control sub-agent model selection. Not checked yet Adds configurable OpenTelemetry OpenTelemetry An open standard and set of tools for collecting traces, metrics and logs from applications. It lets services export observability data without tying instrumentation to one monitoring vendor. logging level. Not checked yet Adds GoosePlatform in AgentConfig and MCP initialization. Not checked yet Adds Gemini CLI streaming support via stream-json events. Not checked yet Exposes context window context window The maximum amount of text, counted in tokens, a model can take in and produce in one request. Everything the model knows about the task has to fit inside it. utilization to the agent via MOIM. Not checked yet Adds auto-submit for recipes that have been accepted. Not checked yet Adds Claude Code permission prompt routing for approve mode. Not checked yet Displays token counts directly for free providers. Not checked yet Adds a TypeScript SDK for ACP extension methods. Not checked yet Adds Bedrock prompt cache support. Not checked yet Removes allows_unlisted_models flag — custom model entry is now always permitted. Not checked yet Displays working directory in the UI. Not checked yet Adds local model settings access from the bottom-bar model menu. Not checked yet
22 details in Everything
└──▷ BREAKING ON UPGRADE
! The allows_unlisted_models flag has been removed; custom model entry is now unconditionally enabled and any configuration relying on this flag will no longer be respected.
Goose v1.0.25 adds Snowflake provider, a full cron scheduler with pause/cancel/edit, LanceDB tool selection, and Claude 4 on Vertex.
└──▷ GET THIS VERSION
copy
$ git clone --branch v1.0.25 https://github.com/block/goose.git
$ git checkout v1.0.25
Adds Snowflake as a supported LLM LLM A large language model, a neural network trained on large amounts of text to recognize patterns and generate language. Applications use it for tasks such as answering, summarizing and writing code. provider. Not checked yet Introduces a cron scheduler for goose-server with schedule creation, editing, pause/unpause, task cancellation, and live run tracking. Not checked yet Supports schedule creation from deep links with comprehensive extension support. Not checked yet Adds MCP MCP Model Context Protocol, a standard for connecting AI applications to external tools and data sources. A compatible server exposes capabilities an assistant can use through a common interface. router skeleton and handling of MCP server notification messages. Not checked yet Enables LanceDB vector-based tool selection for smarter tool routing. Not checked yet
+9 more
show less
Adds GitHub Copilot streaming support, enabling gpt-4.1 and Claude models via Copilot. Not checked yet Adds Claude 4 support in the Vertex provider dropdown with proper context window context window The maximum amount of text, counted in tokens, a model can take in and produce in one request. Everything the model knows about the task has to fit inside it. limits. Not checked yet Enables viewing and editing existing recipes in the Desktop UI. Not checked yet Supports recipe discovery via the GOOSE_RECIPE_PATH environment variable and recipe directory paths. Not checked yet Adds screenshot paste support in the Desktop UI. Not checked yet Enables interactive deletion of sessions from the CLI. Not checked yet Introduces configurable tool_params_max_length setting. Not checked yet Adds Google Drive labels support. Not checked yet Adds a Kotlin function for getting structured outputs structured outputs Forcing a model's response to match a given format or JSON schema, so software can parse it reliably instead of hoping the text is well formed. . Not checked yet
14 details in Everything
└──▷ BREAKING ON UPGRADE
! Tool router environment variable names have been updated to include a goose prefix (e.g., existing env vars without the prefix will no longer be recognized).
Goose v1.25.0 adds macOS sandboxing sandboxing Running code inside a restricted environment that limits what it can access or change. The isolation reduces the impact of mistakes or untrusted code, depending on the sandbox's controls. , unified summon extension, Claude Code streaming, MCP MCP Model Context Protocol, a standard for connecting AI applications to external tools and data sources. A compatible server exposes capabilities an assistant can use through a common interface. support for agentic CLI providers, and SLSA SLSA Supply-chain Levels for Software Artifacts, a framework for assessing how software was built and protecting that process from tampering. Provenance records help verify where a build came from. build provenance provenance Evidence recording where a software artifact came from and how it was built, including its source and build process. It helps a consumer check that the artifact matches a trusted build. .
└──▷ GET THIS VERSION
copy
$ git clone --branch v1.25.0 https://github.com/block/goose.git
$ git checkout v1.25.0
Runs Goose Desktop inside a macOS sandbox for enhanced security isolation. Not checked yet Introduces a unified summon extension, replacing the separate subagent and skills extensions. Not checked yet Integrates MCP Apps UI via AppRenderer from the @mcp-ui/client SDK with fallback request handler support. Not checked yet Enables editing a recipe's model and extensions directly in the desktop GUI GUI A graphical user interface, controls such as windows, buttons and menus. It lets people use software visually rather than entering commands. . Not checked yet Allows recipes to specify their own provider and model via recipe config. Not checked yet
+16 more
show less
Adds streaming output support for the Claude Code CLI provider. Not checked yet Supports dynamic model listing and mid-session model switching for Claude Code. Not checked yet Gemini CLI provider now uses stream-json output and reuses sessions for efficiency. Not checked yet MCP extensions now work with agentic CLI providers (Claude Code, Codex, Gemini CLI). Not checked yet Exposes reasoning_content in the API for reasoning models. Not checked yet Adds support for the extra field in chat completion tool calls for Gemini OpenAI compatibility. Not checked yet Respects standard OpenTelemetry OpenTelemetry An open standard and set of tools for collecting traces, metrics and logs from applications. It lets services export observability data without tying instrumentation to one monitoring vendor. environment variables for exporter selection. Not checked yet Adds model selection support for ACP session/new and session/set_model endpoints. Not checked yet Allows overriding default bat syntax highlighting themes via environment variables. Not checked yet Sets AGENT=goose environment variable automatically for cross-tool compatibility. Not checked yet Generates manpages automatically for goose-cli. Not checked yet Enables the Top Of Mind (tom) platform extension by default. Not checked yet Filters the Apps page to show only standalone Goose Apps. Not checked yet Advertises MCP Apps UI extension capability (via rmcp 0.15.0 upgrade). Not checked yet Supports custom headers for custom OpenAI-compatible providers in the Desktop. Not checked yet Release workflows now include SLSA build provenance attestations for supply-chain security. Not checked yet
21 details in Everything
Goose v1.24.0 adds unified summon extension, MCP MCP Model Context Protocol, a standard for connecting AI applications to external tools and data sources. A compatible server exposes capabilities an assistant can use through a common interface. tools for agentic CLI providers, session search API, and SLSA SLSA Supply-chain Levels for Software Artifacts, a framework for assessing how software was built and protecting that process from tampering. Provenance records help verify where a build came from. provenance provenance Evidence recording where a software artifact came from and how it was built, including its source and build process. It helps a consumer check that the artifact matches a trusted build. .
└──▷ GET THIS VERSION
copy
$ git clone --branch v1.24.0 https://github.com/block/goose.git
$ git checkout v1.24.0
└──▷ TRY IT
Filter long lists instantly when picking models or extensions interactively in the CLI.
New unified summon extension replaces separate subagent and skills extensions with a single streamlined interface. Not checked yet New built-in Top of Mind (tom) platform extension enabled by default for contextual awareness. Not checked yet Claude Code, Codex, and Gemini CLI agentic providers can now use MCP tools. Not checked yet Claude Code gains dynamic model listing and mid-session model switching. Not checked yet Claude Code adds stream-json protocol for persistent sessions. Not checked yet
+28 more
show less
Reasoning content is now returned in the API for reasoning models. Not checked yet Configurable thinking level for Gemini 3 models. Not checked yet Codex now supports image inputs. Not checked yet Google provider adds MessageContent::Image support for user messages. Not checked yet Recipes can now have their model and extensions edited directly from the GUI GUI A graphical user interface, controls such as windows, buttons and menus. It lets people use software visually rather than entering commands. . Not checked yet Recipes can load the provider/model specified inside the recipe config itself. Not checked yet Jinja variables can now be escaped in recipes to pass them through as literal text in prompts. Not checked yet New API endpoint for searching session content. Not checked yet Inline rename for chat sessions in the sidebar. Not checked yet Desktop UI now supports deleting custom providers. Not checked yet Custom headers field added for custom OpenAI-compatible providers in the Desktop UI. Not checked yet Subagent tool calls are now displayed in both the CLI and UI. Not checked yet Type-to-search filtering added to CLI select/multiselect dialogs. Not checked yet AGENT_SESSION_ID environment variable is now exposed to extension child processes. Not checked yet AGENT=goose environment variable set for cross-tool compatibility. Not checked yet Environment variables are now passed to shell commands. Not checked yet Global config option to disable automatic session naming. Not checked yet Port of Context (pctx) support added for Code Mode. Not checked yet Default bat syntax-highlight themes can now be overridden via environment variables. Not checked yet MCP Apps gain Permission Policy support for sandbox iframes. Not checked yet MCP Apps integrate the AppRenderer from the @mcp-ui/client SDK. Not checked yet Upgraded rmcp to 0.15.0 with MCP Apps UI extension capability advertised. Not checked yet Groq provider updated with Preview Models. Not checked yet ACP protocol gains model selection support via session/new and session/set_model. Not checked yet Manpage generation added for goose-cli. Not checked yet SLSA build provenance attestations added to release workflows. Not checked yet Supports building with CUDA CUDA NVIDIA's platform and programming interface for running general computation on its GPUs. Nearly every AI training and inference stack is built on it first. as the candle backend. Not checked yet Proper proxy support on Windows and macOS. Not checked yet
33 details in Everything
Goose v1.23.0 adds WebSocket WebSocket A protocol that upgrades an HTTP connection into a persistent, full-duplex channel, letting a server push data to a client without repeated polling. Cyber tools use it for real-time event streaming and live command output. and Streamable HTTP Streamable HTTP An MCP transport protocol that lets a client send JSON-RPC requests over HTTP POST while the server streams responses back using Server-Sent Events, replacing the older stdio and SSE transports for remote tool communication. transports for ACP, Laminar observability integration, and a CLI flag to skip profile extensions.
└──▷ GET THIS VERSION
copy
$ git clone --branch v1.23.0 https://github.com/block/goose.git
$ git checkout v1.23.0
Adds WebSocket transport support for goose-acp agent communication. Not checked yet Adds Streamable HTTP transport for ACP and goose-acp usage. Not checked yet Integrates Laminar for agent observability and tracing. Not checked yet New CLI flag to skip loading profile extensions at startup. Not checked yet Adds a thinking status hook for surfacing model reasoning state. Not checked yet
+3 more
show less
Adds hourly skill fetching to keep skills automatically up to date. Not checked yet Introduces a platform extension migrator alongside a code mode rename. Not checked yet Adds an Ask AI Discord bot integration. Not checked yet
8 details in Everything
Goose v1.22.0 adds session forking, BERT BERT A transformer encoder trained to understand text using context on both sides of a word. Its representations are useful for search, classification and other language-understanding tasks. -based command injection command injection An attack that makes an application execute operating-system commands supplied through untrusted input. It often happens when input is passed unsafely to a shell. detection, container extensions, and an OSS OSS Open-source software: code whose licence permits anyone to inspect, modify, and redistribute it. Cyber tools build on it to reuse auditable libraries without licensing fees. skills marketplace.
└──▷ GET THIS VERSION
copy
$ git clone --branch v1.22.0 https://github.com/block/goose.git
$ git checkout v1.22.0
└──▷ USE IT
Cap how many turns a recipe's subagent is allowed to run, preventing runaway automation.
yaml
copy
# In your recipe YAML
max_turns: 10
Authenticate to AWS Bedrock using a bearer token bearer token A credential that grants access to whoever presents it, without a separate proof of possession. Protecting the token matters because a stolen copy can be used by someone else. instead of standard AWS credentials.
copy
$ export AWS_BEARER_TOKEN_BEDROCK=<your-token>
goose run --provider bedrock --model anthropic.claude-3-5-sonnet-20241022-v2:0
Adds session forking to branch existing conversations into independent sessions. Not checked yet Introduces BERT ML-based command injection detection in tool calls for prompt injection prompt injection An attack where text the model reads, such as a web page, file or tool result, carries instructions that override the user's. It is the main security risk for agents with tools. defense. Not checked yet Supports running extensions inside containers via new Container Extensions option. Not checked yet Adds an open-source skills marketplace for discovering and adding community skills. Not checked yet Integrates OVHcloud AI as a new model provider. Not checked yet
+15 more
show less
Adds max_turns setting to recipes and subagents to cap agent execution depth. Not checked yet Supports requires_auth flag for configuring custom providers that need no authentication. Not checked yet Adds AWS_BEARER_TOKEN_BEDROCK environment variable support for AWS Bedrock authentication. Not checked yet Supports custom headers with the Anthropic provider. Not checked yet Enables external editor support (vim, helix, etc.) for composing prompts. Not checked yet Adds a built-in Goose Documentation Skill for in-session reference to Goose docs. Not checked yet Adds a Diagnostics Viewer for debugging Goose sessions. Not checked yet Makes skills available in codemode. Not checked yet Adds PR code review functionality. Not checked yet Adds Vibe MCP MCP Model Context Protocol, a standard for connecting AI applications to external tools and data sources. A compatible server exposes capabilities an assistant can use through a common interface. Apps, a new MCP apps integration. Not checked yet Adds collapsible chat navigation for a cleaner desktop UI. Not checked yet Adds new keyboard shortcuts. Not checked yet Adds reduced motion support for accessibility preferences. Not checked yet Creates a default gooseignore file automatically when one is missing. Not checked yet Adds Codex subscription support as a provider option. Not checked yet
20 details in Everything
Goose v1.21.0 adds multi-chat sessions, standalone MCP MCP Model Context Protocol, a standard for connecting AI applications to external tools and data sources. A compatible server exposes capabilities an assistant can use through a common interface. apps, CLI permission prompts, parallel isolated agent sessions, and an MCP server for recipes.
└──▷ GET THIS VERSION
copy
$ git clone --branch v1.21.0 https://github.com/block/goose.git
$ git checkout v1.21.0
Adds multi-chat support to manage multiple concurrent chat sessions. Not checked yet Supports standalone MCP applications with optional prefersBorder UI option. Not checked yet Adds an MCP server for recipes, enabling recipe management via the MCP protocol. Not checked yet Introduces CLI permission prompts so the agent asks for user approval before sensitive actions. Not checked yet Enables parallel sessions with isolated agent state in goose-acp. Not checked yet
+8 more
show less
Adds Flatpak packaging for Linux distributions. Not checked yet Supports user-configurable templates for customizing agent prompts. Not checked yet Adds GCP Vertex model listing with organization policy filtering. Not checked yet Supports optional request params and context limit configuration via GOOSE_PREDEFINED_MODELS. Not checked yet Adds customizable new-line keybinding in the CLI. Not checked yet Standardizes agent-session-id across providers for improved session correlation. Not checked yet Excludes subagent tool from code_execution filtering, enabling subagent delegation in code execution mode. Not checked yet Adds graceful fallback handling when the system keyring is unavailable. Not checked yet
13 details in Everything
Goose v1.20.0 adds ML prompt-injection prompt-injection An attack where text the model reads, such as a web page, file or tool result, carries instructions that override the user's. It is the main security risk for agents with tools. detection, HTTP proxy support, streaming for Gemini and Vertex AI, and a configurable max-tokens env var.
└──▷ GET THIS VERSION
copy
$ git clone --branch v1.20.0 https://github.com/block/goose.git
$ git checkout v1.20.0
└──▷ TRY IT
Cap token usage per session to avoid runaway costs when running Goose against large codebases.
copy
$ GOOSE_MAX_TOKENS=4096 goose run --text "Summarize all TODOs in this repo"
Route Goose's provider API calls through a corporate HTTP proxy for environments that block direct outbound traffic.
copy
$ HTTPS_PROXY=http://proxy.corp.example.com:8080 goose run --text "Scan this project for secrets"
Adds ML-based prompt injection detection to identify and reduce prompt injection attacks, with tuned confidence levels to lower false-positive rates. Not checked yet Supports HTTP proxy via new proxy configuration, enabling Goose to route traffic through corporate or intercepting proxies. Not checked yet Adds GOOSE_MAX_TOKENS environment variable to cap the maximum tokens sent to the model. Not checked yet Adds streaming support for the Google Gemini provider, reducing latency for long responses. Not checked yet Adds streaming support for the Vertex AI provider. Not checked yet
+8 more
show less
Adds retry logic for model fetching from providers, improving resilience against transient API errors. Not checked yet Adds a hotkey to toggle full tool output display in the UI. Not checked yet Adds frameDomains and baseUriDomains Content Security Policy Content Security Policy A browser security mechanism, standardized by the W3C, that lets a web server declare via HTTP headers which sources of scripts, styles, and other resources a page is allowed to load, blocking unauthorized content injection. support for MCP MCP Model Context Protocol, a standard for connecting AI applications to external tools and data sources. A compatible server exposes capabilities an assistant can use through a common interface. Apps, enabling controlled embedding embedding Lists of numbers a model produces to represent the meaning of text, images or code, so similar items sit close together. They power semantic search and RAG. of external domains. Not checked yet Enables changing the working directory and loaded extensions within the same session window without restarting. Not checked yet CLI now surfaces extension load errors directly, making misconfigured extensions immediately visible. Not checked yet Adds spell-check setting to the desktop UI. Not checked yet Hides advanced recipe options under expandable content, reducing UI clutter for common workflows. Not checked yet Caches tools in ExtensionManager for faster code-execution startup. Not checked yet
13 details in Everything
Goose v1.19.0 adds shell completions, OpenAI Codex CLI provider, JSONL JSONL A text format where each line is a self-contained, valid JSON object, making it easy to stream, append, and process large datasets line by line without parsing a full JSON array into memory. streaming, and MCP MCP Model Context Protocol, a standard for connecting AI applications to external tools and data sources. A compatible server exposes capabilities an assistant can use through a common interface. server support from editors like Zed.
└──▷ GET THIS VERSION
copy
$ git clone --branch v1.19.0 https://github.com/block/goose.git
$ git checkout v1.19.0
└──▷ TRY IT
Set up persistent shell completions for zsh so Goose subcommands and flags tab-complete in your terminal.
copy
$ goose completion zsh > ~/.zfunc/_goose
Adds completion subcommand to generate shell completions for bash, zsh, fish, and more. Not checked yet Adds OpenAI Codex CLI as a supported provider. Not checked yet Enables prompt caching for Anthropic models served via Databricks. Not checked yet Honors MCP servers configured in clients like Zed over stdio and HTTP transports. Not checked yet Adds MCP app renderer for richer, interactive UI experiences within MCP sessions. Not checked yet
+2 more
show less
Adds JSONL streaming option for JSON output mode, enabling line-by-line consumption of streamed responses. Not checked yet Uses MCP server names as extension identifiers in code mode for clearer context. Not checked yet
7 details in Everything
└──▷ BREAKING ON UPGRADE
! SSE SSE Server-Sent Events, a browser API and HTTP-based protocol where a server pushes a stream of text events to a client over a single long-lived connection, giving tools a lightweight alternative to WebSockets for one-way real-time updates. transport for MCP is removed as part of the upgrade to rmcp 0.12.0 and sacp 10.0.0; configurations relying on SSE transport will break.
Goose v1.18.0 adds slash commands for prompt/history management, OpenRouter OpenRouter A unified API gateway that routes requests to dozens of LLM providers under one endpoint, letting cyber tools swap or compare models without rewriting provider-specific integration code. & xAI streaming, and skills discovery paths.
└──▷ GET THIS VERSION
copy
$ git clone --branch v1.18.0 https://github.com/block/goose.git
$ git checkout v1.18.0
└──▷ TRY IT
Browse available prompts and run one by name during an interactive session — useful for quickly invoking saved prompt templates.
copy
$ /prompts
/prompt my-security-review-prompt
Inspect a prompt's parameters and metadata before executing it, helpful when you want to understand a recipe's required inputs.
copy
$ /prompt my-security-review-prompt --info
Compact or clear conversation history mid-session to stay within context limits without starting a new session.
copy
$ /compact
# or, to wipe the full history:
/clear
Adds /prompts slash command to list available prompts, /prompt to execute a prompt or inspect it with --info, /compact to compact conversation history, and /clear to clear conversation history. Not checked yet Adds recipe slash command parsing with improved parameter handling. Not checked yet Adds streaming support for OpenRouter and xAI providers. Not checked yet Adds skills auto-discovery from .agents/skills and ~/.config/agent/skills paths. Not checked yet Adds a Goose PR reviewer workflow for automated code review. Not checked yet
+2 more
show less
Adds pricing integration with the canonical model registry. Not checked yet Adds ability to disable the auto-update feature for the CLI. Not checked yet
7 details in Everything
Goose v1.17.0 adds Inception provider, MCP MCP Model Context Protocol, a standard for connecting AI applications to external tools and data sources. A compatible server exposes capabilities an assistant can use through a common interface. elicitation, external server support, OpenAI streaming, and new built-in recipes.
└──▷ GET THIS VERSION
copy
$ git clone --branch v1.17.0 https://github.com/block/goose.git
$ git checkout v1.17.0
Adds Code Mode MCP platform extension for enhanced code execution capabilities via tool calling tool calling A model's ability to answer with a structured request to run a named function or API, which the application executes and returns. It is how agents take actions. . Not checked yet Supports MCP elicitation, enabling interactive prompts during tool execution. Not checked yet Enables connection to an external goosed server backend. Not checked yet Adds Inception provider, bringing Mercury model support to Goose. Not checked yet Adds streaming support for the OpenAI responses API. Not checked yet
+5 more
show less
Introduces canonical (standardized) model definitions across providers. Not checked yet Adds search bar and filtering to the recipes view. Not checked yet Enables automatic provider detection from API key during onboarding. Not checked yet Adds built-in DataHub recipe for searching and understanding data in DataHub. Not checked yet Adds built-in Flutter PR Code Review recipe for reviewing Flutter pull requests. Not checked yet
10 details in Everything
Goose v1.16.0 adds native terminal support, remote access, slash commands, custom headers, and a new /agent/stop API endpoint.
└──▷ GET THIS VERSION
copy
$ git clone --branch v1.16.0 https://github.com/block/goose.git
$ git checkout v1.16.0
└──▷ TRY IT
Invoke Goose directly from your native terminal without launching the desktop UI.
copy
$ @goose explain the CVEs in this repo's dependencies
Stop a running agent via the new API endpoint in a headless or CI environment.
copy
$ curl -X POST http://localhost:<port>/agent/stop
Adds @goose native terminal support, enabling Goose interaction directly from your system terminal. Not checked yet Adds Goose remote access capability for controlling Goose from external clients. Not checked yet Adds slash commands in the chat interface for quick in-session actions. Not checked yet Adds custom headers and toggle keyring CLI options for provider configuration. Not checked yet Adds /agent/stop API endpoint and makes max active agents configurable. Not checked yet
+13 more
show less
Adds support for loading sessions in ACP (Agent Communication Protocol). Not checked yet Enables recipe deeplink parameters for pre-population of recipe inputs. Not checked yet Adds SKILLS.md support for Claude compatibility. Not checked yet Adds math and science symbology rendering via KaTeX. Not checked yet Adds markdown export from URL. Not checked yet Adds Firefox sidebar contextual information support. Not checked yet Adds automatic update installation. Not checked yet Adds MCP MCP Model Context Protocol, a standard for connecting AI applications to external tools and data sources. A compatible server exposes capabilities an assistant can use through a common interface. -UI proxy for better MCP message handling. Not checked yet Adds scheduler information to diagnostics output. Not checked yet Persists provider name and model config within the session. Not checked yet Adds extension selector behind an ALPHA feature flag. Not checked yet Adds multiple file popup support for simultaneous file interactions. Not checked yet Supports onboarding auto-detection of provider from API key. Not checked yet
18 details in Everything
Goose v1.15.0 adds Grok provider support, Supabase MCP MCP Model Context Protocol, a standard for connecting AI applications to external tools and data sources. A compatible server exposes capabilities an assistant can use through a common interface. , recipe security scanning, and ACP session management.
└──▷ GET THIS VERSION
copy
$ git clone --branch v1.15.0 https://github.com/block/goose.git
$ git checkout v1.15.0
Adds Grok as a supported AI provider. Not checked yet Adds Supabase MCP server to the extension registry. Not checked yet Adds security scanning of recipes at save time to catch malicious or unsafe content. Not checked yet Adds SessionManager integration for ACP (Agent Communication Protocol) sessions. Not checked yet Adds MCP UI proxy with improved message handling for extension interactions. Not checked yet
+2 more
show less
Reads paths from both interactive and login shells, improving environment compatibility. Not checked yet Adds provider error proxy for simulating provider errors during development and testing. Not checked yet
7 details in Everything
Goose v1.14.0 adds Mistral AI and GitHub Copilot providers, platform tools via CLI, and env-var substitution in auth config.
└──▷ GET THIS VERSION
copy
$ git clone --branch v1.14.0 https://github.com/block/goose.git
$ git checkout v1.14.0
Adds Mistral AI as a new LLM LLM A large language model, a neural network trained on large amounts of text to recognize patterns and generate language. Applications use it for tasks such as answering, summarizing and writing code. provider option. Not checked yet Supports newer GitHub Copilot AI model versions as a provider. Not checked yet Enables access to platform tools through the CLI. Not checked yet Supports dynamic environment variable substitution in authentication configuration blocks. Not checked yet Adds a macOS dock menu with a New Window option for better desktop integration. Not checked yet
+1 more
show less
Adds a pending extension indicator in the extension panel for visual loading feedback. Not checked yet
6 details in Everything
Goose v1.13.0 adds Mistral AI provider, newer GitHub Copilot models, platform tools in CLI, and env-var auth config.
└──▷ GET THIS VERSION
copy
$ git clone --branch v1.13.0 https://github.com/block/goose.git
$ git checkout v1.13.0
Adds Mistral AI as a supported LLM LLM A large language model, a neural network trained on large amounts of text to recognize patterns and generate language. Applications use it for tasks such as answering, summarizing and writing code. provider. Not checked yet Supports newer GitHub Copilot models. Not checked yet Exposes platform tools in the CLI (previously desktop-only). Not checked yet Enables environment variables in auth configuration. Not checked yet Adds Ctrl-C cancellation of in-flight AI requests from the CLI. Not checked yet
+2 more
show less
Adds right-click dock icon for opening a new window on macOS. Not checked yet Shows a loading indicator for extensions while they initialize. Not checked yet
7 details in Everything
└──▷ BREAKING ON UPGRADE
! The JetBrains integration has been removed; any existing JetBrains-based workflows will no longer function after upgrading.
Goose v1.12.0 adds system diagnostics, extension search, code snippet saving, and two new built-in recipes.
└──▷ GET THIS VERSION
copy
$ git clone --branch v1.12.0 https://github.com/block/goose.git
$ git checkout v1.12.0
Adds system diagnostics tools for inspecting Goose's runtime environment. Not checked yet Adds extension search functionality to quickly find and manage extensions. Not checked yet Adds code snippet formatting and saving directly from conversations. Not checked yet Introduces a Test Coverage Optimizer built-in recipe. Not checked yet Introduces a Smart Task Organizer built-in recipe. Not checked yet
+7 more
show less
Adds a schedule button for recipe entries, enabling scheduling from the recipe list UI. Not checked yet Auto-creates recipe directories when saving a recipe to a new path. Not checked yet Adds UI parameter input controls for configuring recipe parameters interactively. Not checked yet Redesigns the extension management system. Not checked yet Adds progress tracking with trend indicators. Not checked yet Improves prompt caching with stable system prompts for more efficient provider usage. Not checked yet Adds provider request logging for visibility into outbound LLM LLM A large language model, a neural network trained on large amounts of text to recognize patterns and generate language. Applications use it for tasks such as answering, summarizing and writing code. requests. Not checked yet
12 details in Everything
Goose v1.11.0 adds session search, new built-in recipes, Apify MCP MCP Model Context Protocol, a standard for connecting AI applications to external tools and data sources. A compatible server exposes capabilities an assistant can use through a common interface. integration, voice dictation, and declarative providers.
└──▷ GET THIS VERSION
copy
$ git clone --branch v1.11.0 https://github.com/block/goose.git
$ git checkout v1.11.0
Introduces a Code Documentation Generator built-in recipe. Not checked yet Introduces a CI/CD Pipeline built-in recipe. Not checked yet Introduces a Daily Standup Report Generator built-in recipe. Not checked yet Adds a self-test recipe for validating Goose installations. Not checked yet Adds an API Documentation Generator prompt. Not checked yet
+9 more
show less
Adds Apify MCP integration with supporting documentation. Not checked yet Adds a flag to enable ElevenLabs voice dictation. Not checked yet Introduces declarative provider configuration. Not checked yet Sorts providers alphabetically for consistent discovery. Not checked yet Adds a Copy as Markdown button to documentation pages. Not checked yet Enables starting an agent session directly from a recipe ID or deeplink. Not checked yet Improves Rust analysis output for the analyze tool. Not checked yet Adds Firecrawl MCP tutorial and integration documentation. Not checked yet Expands Ollama integration with a modified default model and additional models. Not checked yet
14 details in Everything
Goose v1.10.0 adds prompt injection prompt injection An attack where text the model reads, such as a web page, file or tool result, carries instructions that override the user's. It is the main security risk for agents with tools. detection UI, GOOSE_DEBUG support, custom lead/worker models, AWS Bearer Token Bearer Token A credential that grants access to whoever presents it, without a separate proof of possession. Protecting the token matters because a stolen copy can be used by someone else. auth, and more.
└──▷ GET THIS VERSION
copy
$ git clone --branch v1.10.0 https://github.com/block/goose.git
$ git checkout v1.10.0
└──▷ TRY IT
Enable verbose debug output when troubleshooting CLI sessions or extension behavior.
copy
$ GOOSE_DEBUG=1 goose session start
List only the most recent sessions under a specific path to audit or resume recent work.
copy
$ goose session list --path /projects/myapp --limit 10
Open a saved recipe directly in the Goose desktop app from the CLI.
copy
$ goose recipe open <path>
Adds prompt injection detection settings UI with updated logging for security-aware deployments. Not checked yet New GOOSE_DEBUG environment variable for CLI debug output. Not checked yet Adds path and limit flags to the session list subcommand for filtering session history. Not checked yet New subcommand for opening recipes directly in the desktop app. Not checked yet Supports setting custom models for lead and worker agents in multi-model workflows. Not checked yet
+22 more
show less
Adds grok-code-fast-1 model support for the xAI provider. Not checked yet Adds claude-sonnet-4-5 model support. Not checked yet Enables AWS Bearer Token (AWS_BEARER_TOKEN_BEDROCK) authentication for Bedrock runtime. Not checked yet Parses and respects retryDelay from Google API RateLimitExceeded errors for smarter rate-limit handling. Not checked yet Adds elapsed time display to CLI session output. Not checked yet Lazy infinite scroller for session history view improves navigation of large session lists. Not checked yet Supports headers in extension deeplinks. Not checked yet Allows empty API key when registering a custom provider. Not checked yet Shell errors thrown on interactive commands to surface blocked interactions explicitly. Not checked yet AgentManager foundation enables unified subagent execution and concurrent session creation. Not checked yet Persists dynamic extension config so recipe sessions with extensions can be resumed. Not checked yet Displays extension install notes in the Add custom extension form. Not checked yet Adds dev and alpha environment indicator in the UI. Not checked yet Using --resume with --name now also accepts session IDs. Not checked yet CLI now returns a non-zero exit code when a session ends with an error. Not checked yet Goose can now tell the system what shell it is using. Not checked yet New Code Documentation Migrator, Smart Meeting Assistant, Code Review Mentor, Smart Task Organizer, advanced data analysis pipeline, advanced software project generator, Messy Column Fixer, and OpenMetadata recipes added to the prompt/recipe library. Not checked yet New multi-channel researcher and content amplification prompts added to the prompt library. Not checked yet New analyze tool hint for focus mode when used on file paths. Not checked yet Screen access capability added (feat: Use the screen, goose). Not checked yet CLI web auth token support added. Not checked yet Session manager added for improved session lifecycle management. Not checked yet
27 details in Everything
Goose v1.9.2 adds Claude Sonnet 4.5 model support and macOS Tahoe compatibility.
└──▷ GET THIS VERSION
copy
$ git clone --branch v1.9.2 https://github.com/block/goose.git
$ git checkout v1.9.2
Adds claude-sonnet-4-5 as a supported model. Not checked yet Upgrades Electron runtime for macOS Tahoe compatibility. Not checked yet
2 details in Everything
Goose v1.9.0 adds prompt injection prompt injection An attack where text the model reads, such as a web page, file or tool result, carries instructions that override the user's. It is the main security risk for agents with tools. detection, auto-compaction, an analyze tool, streaming for Tetrate ARS, and richer recipe management in the UI.
└──▷ GET THIS VERSION
copy
$ git clone --branch v1.9.0 https://github.com/block/goose.git
$ git checkout v1.9.0
Adds prompt injection detection (pattern-matching) to flag malicious inputs before they reach the model. Not checked yet Automatically compacts sessions for streaming providers when context length context length The maximum amount of text, counted in tokens, a model can take in and produce in one request. Everything the model knows about the task has to fit inside it. is exceeded, with an offer to summarize or clear overly long conversations. Not checked yet New analyze tool added to the Developer extension for code and file analysis. Not checked yet Adds streaming support to the Tetrate Agent Router Service (ARS) provider. Not checked yet Supports importing file contents from recipe file input-type parameters. Not checked yet
+8 more
show less
Desktop UI now supports uploading JSON recipe files directly. Not checked yet Recipe schema validation added to the UI to catch malformed recipes before execution. Not checked yet UI recipe activities now support parameter replacement and always appear at the top of chat when a recipe is active. Not checked yet Adds Message Metadata for Visibility Control, enabling fine-grained control over which messages are shown. Not checked yet I Ching MCP MCP Model Context Protocol, a standard for connecting AI applications to external tools and data sources. A compatible server exposes capabilities an assistant can use through a common interface. server added to the extension catalog. Not checked yet Upgraded agent-client-protocol (ACP) with improved tool experience and nicer tool name formatting. Not checked yet Recipe title field added to the import form. Not checked yet UI filters invalid, duplicate, and unused parameters from recipes automatically. Not checked yet
13 details in Everything
└──▷ BREAKING ON UPGRADE
! The --path flag has been removed to avoid session directory validation issues.
Goose v1.8.0 adds ACP support, multi-model switching, streaming agent routing, and YAML recipe imports.
└──▷ GET THIS VERSION
copy
$ git clone --branch v1.8.0 https://github.com/block/goose.git
$ git checkout v1.8.0
Adds Agent Client Protocol (ACP) support for interoperability with ACP-compatible agents. Not checked yet Supports multi-model/provider configuration with automatic switching between providers. Not checked yet Enables streaming support in the Tetrate Agent Router. Not checked yet Adds configurable Bedrock retry settings. Not checked yet Supports importing YAML recipes directly in the Desktop app. Not checked yet
+9 more
show less
Enables reading local file inputs inside recipes. Not checked yet Adds message metadata for visibility control over what is shown in the UI. Not checked yet Enables deleting sessions from the history list in the Desktop UI. Not checked yet Tracks sessions per agent for better session management. Not checked yet Makes the auto-compact threshold configurable via the UI. Not checked yet Shows enabled extensions first in the extensions list. Not checked yet Adds a unified diff view in the developer editor. Not checked yet Aligns recipe metadata metrics across scheduler, UI, and CLI. Not checked yet Upgrades to RMCP 0.6.2. Not checked yet
14 details in Everything
└──▷ BREAKING ON UPGRADE
! Deprecated Claude 3.5 models have been removed.
Goose v1.7.0 adds Android/Termux computer control, extension malware scanning, auto-visualization of structured data, and dynamic provider model querying.
└──▷ GET THIS VERSION
copy
$ git clone --branch v1.7.0 https://github.com/block/goose.git
$ git checkout v1.7.0
Adds computer control support on Android devices via Termux. Not checked yet Adds extension malware scanning to detect malicious code before installation. Not checked yet New MCP MCP Model Context Protocol, a standard for connecting AI applications to external tools and data sources. A compatible server exposes capabilities an assistant can use through a common interface. -UI Autovisualiser automatically renders structured data returned in sessions. Not checked yet UI now dynamically queries providers for their supported models at runtime. Not checked yet Re-introduces session sharing between users. Not checked yet
+5 more
show less
Supports port numbers in custom OpenAI-compatible provider configurations. Not checked yet Adds support for reading OTLP OTLP OpenTelemetry Protocol, a standard way to send traces, metrics and logs between applications, collectors and monitoring systems. It lets telemetry move across compatible tools. configuration from config files and environment variables. Not checked yet Enables loading of nested hint files for better project organization. Not checked yet Adds Tetrate Agent Router Service as a new provider in the registry. Not checked yet Adds scroll state support for improved navigation in the chat session list. Not checked yet
10 details in Everything
Goose v1.6.0 adds mTLS mTLS Mutual TLS, an encrypted connection where both client and server present certificates to prove their identity. It helps services verify who they are communicating with. for all providers, PKCE PKCE Proof Key for Code Exchange, an OAuth safeguard that binds an authorization code to the client that requested it. An intercepted code cannot be redeemed without the client's matching verifier. auth, message queue with interruption handling, and in-place message editing.
└──▷ GET THIS VERSION
copy
$ git clone --branch v1.6.0 https://github.com/block/goose.git
$ git checkout v1.6.0
└──▷ TRY IT
Allow Goose to automatically load project-level agent instructions without any extra flags.
copy
$ # Place an AGENTS.md file in your project root; Goose reads it by default on startup
echo '# Agent Instructions
Always use TypeScript.' > AGENTS.md
goose session
Adds mTLS support across all providers for mutual-authentication to AI backends. Not checked yet Adds PKCE support for the Tetrate Agent Router Service authentication flow. Not checked yet New message queue system with interruption handling for more reliable agent interactions. Not checked yet Providers can now configure a fast model specifically for summarization tasks. Not checked yet Supports concurrent extension startup, reducing initialization latency. Not checked yet
+6 more
show less
Desktop app alerts when suspicious Unicode characters are detected in a Recipe. Not checked yet Reads AGENTS.md by default as a context file, bringing project-level agent instructions into scope automatically. Not checked yet MCP MCP Model Context Protocol, a standard for connecting AI applications to external tools and data sources. A compatible server exposes capabilities an assistant can use through a common interface. developer extension now accepts -1 for insert_line number to support end-of-file insertions. Not checked yet Adds a Nix flake Nix flake A Nix packaging construct that pins all dependencies in a lockfile and exposes a standardized interface, giving cyber tools fully reproducible builds across any machine running the Nix package manager. for reproducible Goose builds. Not checked yet Upgrades @mcp-ui/client package with improved UI message handling. Not checked yet Adds remark-breaks plugin to preserve single newlines in rendered markdown. Not checked yet
11 details in Everything
└──▷ BREAKING ON UPGRADE
! The session sharing feature has been removed; any workflows relying on it will no longer function.
Goose v1.5.0 adds cursor-agent CLI provider, OAuth keyring persistence for MCP MCP Model Context Protocol, a standard for connecting AI applications to external tools and data sources. A compatible server exposes capabilities an assistant can use through a common interface. , and per-recipe tool limits.
└──▷ GET THIS VERSION
copy
$ git clone --branch v1.5.0 https://github.com/block/goose.git
$ git checkout v1.5.0
Adds cursor-agent as a new CLI provider option. Not checked yet Persists OAuth credentials to the system keyring for MCP integrations, eliminating repeated auth prompts. Not checked yet Supports per-recipe configuration to limit which tools are available during a recipe run. Not checked yet Improves model selection UX in the CLI. Not checked yet Speeds up extension loading by parallelizing the load process. Not checked yet
+3 more
show less
Skips fetching pricing information when GOOSE_CLI_SHOW_COST is false, reducing unnecessary network calls. Not checked yet Supports loading recipe deeplinks in a single window when the app is closed. Not checked yet Integrates token counting reliability improvements with summarization. Not checked yet
8 details in Everything
Goose v1.3.0 adds OAuth CLI support, file completion, session description editing, client disconnect detection, and Linux build improvements.
└──▷ GET THIS VERSION
copy
$ git clone --branch v1.3.0 https://github.com/block/goose.git
$ git checkout v1.3.0
Adds OAuth provider abstraction for CLI configuration, enabling authenticated provider setup from the command line. Not checked yet Adds file path tab-completion in the CLI for faster input. Not checked yet Enables editing of session descriptions directly in the UI. Not checked yet Detects client disconnects and automatically cancels in-flight tool calls. Not checked yet Adds StreamableHTTP OAuth support via RMCP. Not checked yet
+7 more
show less
Bumps auto-compact threshold to 80%, reducing premature context compaction context compaction Condensing an AI conversation or working history to free space in the model's context window. A summary preserves useful information, but details can be lost. . Not checked yet Token counting in auto-compact now uses provider metadata for more accurate limits. Not checked yet Adds GPT-5 context limit support. Not checked yet Suppresses ANSI escape codes automatically when output is piped. Not checked yet Adds comprehensive Linux build support with cross-compilation in bundle workflow. Not checked yet Adds additional telemetry metrics with UI implementation. Not checked yet Window title now reflects the current active operation. Not checked yet
12 details in Everything
Goose v1.2.0 adds parallel sub-recipe execution, LiteLLM provider, recipe retry logic, wakelock, and more.
└──▷ GET THIS VERSION
copy
$ git clone --branch v1.2.0 https://github.com/block/goose.git
$ git checkout v1.2.0
Adds experimental support for running sub-recipes multiple times in parallel for concurrent task execution. Not checked yet Adds LiteLLM provider with automatic prompt caching support. Not checked yet Adds recipe retry logic with configurable success criteria so recipes can self-heal on failure. Not checked yet Adds wakelock feature to prevent the system from sleeping while Goose is working. Not checked yet Adds inline Python extension support. Not checked yet
+20 more
show less
Adds GitHub Copilot provider ability to fetch supported models. Not checked yet Adds OpenRouter OpenRouter A unified API gateway that routes requests to dozens of LLM providers under one endpoint, letting cyber tools swap or compare models without rewriting provider-specific integration code. as an out-of-the-box provider experience for new Goose installations. Not checked yet Adds ability for users to view and edit their non-secret configuration values. Not checked yet Adds support for configuring the hints filename(s) via configuration. Not checked yet Adds WaitingForUserInput state to the agent loop. Not checked yet Adds an option to disable random thinking messages. Not checked yet Adds an option to summarize the chat when an error is triggered (autocompact + one-shot summarization). Not checked yet Adds environment variable to set Ollama request timeout. Not checked yet Adds subagent independent extension manager and per-subagent turn count with a simpler agent loop. Not checked yet Adds consolidated subagent execution for dynamic tasks. Not checked yet Adds recipe install warning before accepting untrusted recipes. Not checked yet Adds file picker for scheduling recipes that defaults to the recipe directory. Not checked yet Adds MCP MCP Model Context Protocol, a standard for connecting AI applications to external tools and data sources. A compatible server exposes capabilities an assistant can use through a common interface. UI integration. Not checked yet Adds mouse back navigation button support on the Settings screen. Not checked yet Adds extension name tooltips on tool banners. Not checked yet Adds goose info -v for stable verbose info output. Not checked yet Updates Groq models for better tool-calling tool-calling A model's ability to answer with a structured request to run a named function or API, which the application executes and returns. It is how agents take actions. support. Not checked yet Adds HTTP 429-style retry handling for Anthropic API HTTP 529 overloaded responses in the GCP Vertex AI provider. Not checked yet Enables streaming for the Anthropic first-party provider. Not checked yet Recipe runs now prompt for missing extension secrets instead of silently failing. Not checked yet
25 details in Everything
Goose v1.1.4 adds chat summarization on error and enhanced loading animations with thinking icons and flying bird.
└──▷ GET THIS VERSION
copy
$ git clone --branch v1.1.4 https://github.com/block/goose.git
$ git checkout v1.1.4
Adds option to summarize the chat session when an error is triggered. Not checked yet Enhances loading states with thinking icons and a flying bird animation. Not checked yet
2 details in Everything
Goose v1.1.0 ships a redesigned desktop UI with sidebar and settings tabs, new recipe list command, glob/grep file search tools, and Windows CLI installer.
└──▷ GET THIS VERSION
copy
$ git clone --branch v1.1.0 https://github.com/block/goose.git
$ git checkout v1.1.0
└──▷ TRY IT
List all available recipes to quickly discover and reference them before running one.
Clear the persisted session history mid-session to start fresh without restarting Goose.
Adds recipe list subcommand to enumerate available recipes from the CLI. Not checked yet New /clear command in CLI sessions clears the persisted session file. Not checked yet Adds glob search and grep tools to improve file search capabilities. Not checked yet Implements OpenAI streaming support for faster, real-time LLM LLM A large language model, a neural network trained on large amounts of text to recognize patterns and generate language. Applications use it for tasks such as answering, summarizing and writing code. responses. Not checked yet Complete redesign of the Goose desktop UI with a sidebar and settings tabs. Not checked yet
+1 more
show less
Adds download_cli.ps1 PowerShell installer script for Windows users. Not checked yet
6 details in Everything
Goose v1.0.36 adds streaming LLM LLM A large language model, a neural network trained on large amounts of text to recognize patterns and generate language. Applications use it for tasks such as answering, summarizing and writing code. responses, cost estimation, OpenRouter OpenRouter A unified API gateway that routes requests to dozens of LLM providers under one endpoint, letting cyber tools swap or compare models without rewriting provider-specific integration code. model discovery, vi edit mode, and new provider/model CLI flags.
└──▷ GET THIS VERSION
copy
$ git clone --branch v1.0.36 https://github.com/block/goose.git
$ git checkout v1.0.36
└──▷ TRY IT
Pick a specific provider and model on the fly without touching your config — useful when testing a new model in a one-off session.
copy
$ goose run --provider openai --model gpt-4o <session-or-prompt>
Streams LLM responses in real time instead of waiting for full completion. Not checked yet Adds per-provider cost estimation in the Goose CLI so practitioners can compare spend before committing to a provider. Not checked yet Fetches OpenRouter's supported model list dynamically inside goose configure. Not checked yet Exposes AZURE_OPENAI_API_KEY as a visible, configurable parameter in the config system. Not checked yet Adds --provider and --model CLI options to the goose run command for one-off provider/model selection. Not checked yet
+5 more
show less
Adds a Streamable HTTP Streamable HTTP An MCP transport protocol that lets a client send JSON-RPC requests over HTTP POST while the server streams responses back using Server-Sent Events, replacing the older stdio and SSE transports for remote tool communication. CLI flag for HTTP-based streaming transport. Not checked yet Supports vi edit mode via rustyline configuration for CLI line editing. Not checked yet Allows Ollama to be used with non-tool (chat-only) models. Not checked yet Adds structured output structured output Forcing a model's response to match a given format or JSON schema, so software can parse it reliably instead of hoping the text is well formed. support in Goose CLI and Goose Desktop. Not checked yet Alphabetizes extensions in the UI for easier navigation. Not checked yet
10 details in Everything
Goose v1.0.32 adds OAuth 2.0 OAuth 2.0 An open authorization framework (IETF RFC 6749) that lets a service grant limited API access to a third party without sharing credentials, giving cyber tools a standard way to authenticate against external platforms. for MCP MCP Model Context Protocol, a standard for connecting AI applications to external tools and data sources. A compatible server exposes capabilities an assistant can use through a common interface. , streamable-HTTP transport, JSON schema recipe validation, and sub-recipes from GitHub.
└──▷ GET THIS VERSION
copy
$ git clone --branch v1.0.32 https://github.com/block/goose.git
$ git checkout v1.0.32
└──▷ TRY IT
Validate a recipe file against its JSON schema to catch structural errors before running it.
copy
$ goose recipe validate <path-to-recipe.yaml>
Adds fuzzy file search functionality for navigating files. Not checked yet Adds JSON schema validation to the goose recipe validate CLI command. Not checked yet Introduces max_turns setting for the agent to cap autonomous runs without user input. Not checked yet Enables structured output structured output Forcing a model's response to match a given format or JSON schema, so software can parse it reliably instead of hoping the text is well formed. support in recipes, usable from both CLI and GUI GUI A graphical user interface, controls such as windows, buttons and menus. It lets people use software visually rather than entering commands. . Not checked yet GUI now renders structured output defined in recipes. Not checked yet
+10 more
show less
Enables running sub-recipes sourced directly from GitHub. Not checked yet Adds native OAuth 2 OAuth 2 An open authorization framework (RFC 6749) that lets a service grant limited access to a user's resources on another service without exposing credentials, giving cyber tools a standard way to authenticate against third-party APIs. .0 authentication support to the MCP client. Not checked yet Adds streamable-HTTP transport support across backend, desktop, and CLI. Not checked yet Supports recipe parameters in the Goose desktop app. Not checked yet Adds a close button (X) to toast notifications. Not checked yet Adds support for the Escape key to dismiss the settings menu. Not checked yet Adds Playwright MCP server to the extensions list. Not checked yet Adds /extension path for extension installation. Not checked yet Improves UX for saving recipes. Not checked yet Prioritizes path suffix when truncating long paths in the desktop header. Not checked yet
15 details in Everything
└──▷ BREAKING ON UPGRADE
! GitHub Copilot Provider has been temporarily removed; any setup relying on it will no longer function.
Goose v1.0.31 adds sub-recipe chaining from the CLI, async token counting, and Claude 4 model support.
└──▷ GET THIS VERSION
copy
$ git clone --branch v1.0.31 https://github.com/block/goose.git
$ git checkout v1.0.31
Supports passing additional sub-recipes via the command line for composable recipe workflows. Not checked yet Adds Claude 4 models as available provider options. Not checked yet Implements async token counter with network resilience and performance optimizations. Not checked yet Allows use of /dev/null for no-session mode, enabling fully stateless runs. Not checked yet
4 details in Everything
Goose v1.0.30 adds voice dictation, subagents, CLI providers for Claude/Gemini, cost tracking, and broader cron schedule support.
└──▷ GET THIS VERSION
copy
$ git clone --branch v1.0.30 https://github.com/block/goose.git
$ git checkout v1.0.30
└──▷ TRY IT
Suppress all non-essential output when running a Goose session in CI or scripted pipelines.
copy
$ goose run --quiet --instructions 'Summarize the repo structure'
Resume a session by its ID using the new short alias instead of --name.
copy
$ goose session --resume --id my-session-id
Adds voice dictation support via OpenAI Whisper and ElevenLabs integration. Not checked yet Introduces subagents capability, enabling Goose to orchestrate multi-agent workflows. Not checked yet New CLI providers for Claude Code and Gemini, expanding model backend options. Not checked yet Adds comprehensive cost tracking display for LLM LLM A large language model, a neural network trained on large amounts of text to recognize patterns and generate language. Applications use it for tasks such as answering, summarizing and writing code. usage, with a flag to toggle visibility. Not checked yet Wider cron schedule support for the built-in scheduler. Not checked yet
+21 more
show less
Adds optional view_range parameter to the text editor tool and text editor insert tool. Not checked yet New sub recipe tools for composing and reusing recipe components. Not checked yet Lists Groq-supported models from the CLI during model configuration. Not checked yet Lists Databricks-supported models with fuzzy search during model configuration. Not checked yet Adds --id alias for --name parameter in goose session --resume command. Not checked yet Adds --quiet / -q flag to goose run for suppressing output. Not checked yet Adds scheduler type setting for controlling how scheduled tasks are dispatched. Not checked yet Adds ability to create a new directory directly from the working directory selection UI. Not checked yet Drag-and-drop support for opening sessions from Finder in the desktop UI. Not checked yet Adds chain-of-thought chain-of-thought Intermediate reasoning a model writes out before its final answer. Reasoning models generate it automatically and it is often billed as extra output tokens. panel displayed above assistant messages in the UI. Not checked yet Supports optional fast edit models for accelerated editing operations. Not checked yet Adds lead-worker model selection and real-time model display in the GUI GUI A graphical user interface, controls such as windows, buttons and menus. It lets people use software visually rather than entering commands. . Not checked yet Updates vector tool strategy to read vector DB path from an environment variable. Not checked yet Updates Google Gemini models to the latest available versions. Not checked yet Adds xAI Provider support for Grok models. Not checked yet Platform tool now allows Goose to manage its own schedule autonomously. Not checked yet Recipe Library feature for discovering and reusing recipes. Not checked yet Adds a setting for the quit confirmation dialog. Not checked yet Richer tool call UI messages for better visibility into agent actions. Not checked yet Improves config file editing with recovery fallback mechanisms. Not checked yet Native Windows CLI build support. Not checked yet
26 details in Everything
Goose v1.0.29 adds a system prompt parameter to run, a /clear command, alphabetized extensions, and a confirmation dialog for unsaved changes.
└──▷ GET THIS VERSION
copy
$ git clone --branch v1.0.29 https://github.com/block/goose.git
$ git checkout v1.0.29
└──▷ TRY IT
Inject a custom system prompt when running a Goose recipe or session non-interactively, useful for enforcing persona or constraints in CI pipelines.
copy
$ goose run --system "You are a security-focused code reviewer. Flag any use of eval() or exec()." --recipe review.yaml
Clear accumulated context during a long session to reduce token usage or start a fresh task without restarting Goose.
Adds --system (system prompt) parameter to the goose run command, enabling custom system-level instructions per run. Not checked yet Adds /clear command to clear the Goose context mid-session. Not checked yet Alphabetizes extensions listing in the Goose CLI for easier navigation. Not checked yet Adds confirmation dialog for unsaved changes when closing the extension modal in the UI. Not checked yet Adds a Help & Feedback section in App Settings. Not checked yet
+2 more
show less
Adds devcontainer support for development environments. Not checked yet Ensures a newline is written at the end of file writes. Not checked yet
7 details in Everything
Goose v1.0.28 adds SageMaker TGI support, an LLM LLM A large language model, a neural network trained on large amounts of text to recognize patterns and generate language. Applications use it for tasks such as answering, summarizing and writing code. tool router, recipe settings, and desktop auto-update.
└──▷ GET THIS VERSION
copy
$ git clone --branch v1.0.28 https://github.com/block/goose.git
$ git checkout v1.0.28
Adds SageMaker TGI as a supported provider backend. Not checked yet New LLM tool selector for the tool router, enabling AI-driven routing of tool calls. Not checked yet Tool router now includes extension names in vector DB and search tool metadata. Not checked yet Goose recipes now support configurable settings. Not checked yet Adds auto-update functionality to the Goose desktop app, gated by an UPDATES_ENABLED flag. Not checked yet
+2 more
show less
CLI no longer halts startup when one or more MCP MCP Model Context Protocol, a standard for connecting AI applications to external tools and data sources. A compatible server exposes capabilities an assistant can use through a common interface. extensions fail to load. Not checked yet Temporal disabled by default, with dynamic port selection when enabled. Not checked yet
7 details in Everything
Goose v1.0.26 adds session export, a temporal scheduler, lead/worker models, Snowflake provider, and Gemini 2.5 support.
└──▷ GET THIS VERSION
copy
$ git clone --branch v1.0.26 https://github.com/block/goose.git
$ git checkout v1.0.26
└──▷ TRY IT
Set a custom context management strategy to control how Goose handles long conversations.
copy
$ goose config set GOOSE_CONTEXT_STRATEGY summarize
Load recipes from a custom directory by setting GOOSE_RECIPE_PATH before running.
copy
$ GOOSE_RECIPE_PATH=/team/recipes goose run --recipe my-recipe
New export CLI command converts sessions to Markdown for sharing or archiving. Not checked yet Adds a temporal scheduler with schedule creation, editing, pause/unpause, live tracking, and task cancellation for scheduled jobs. Not checked yet New lead/worker model architecture enables multi-model agent workflows. Not checked yet New goose web command provides a local web-based terminal alternative. Not checked yet Adds Snowflake as a new LLM LLM A large language model, a neural network trained on large amounts of text to recognize patterns and generate language. Applications use it for tasks such as answering, summarizing and writing code. provider. Not checked yet
+22 more
show less
Adds support for Gemini 2.5 Flash Preview and Pro Preview models. Not checked yet Adds Claude 4 to the Vertex AI provider dropdown with context window context window The maximum amount of text, counted in tokens, a model can take in and produce in one request. Everything the model knows about the task has to fit inside it. limit support. Not checked yet New GOOSE_CONTEXT_STRATEGY CLI config setting controls context management behavior. Not checked yet CLI now shows active context length context length The maximum amount of text, counted in tokens, a model can take in and produce in one request. Everything the model knows about the task has to fit inside it. during sessions. Not checked yet New system prompt override support in goose-llm. Not checked yet Supports deep link-based schedule creation with comprehensive extension support. Not checked yet Adds MCP MCP Model Context Protocol, a standard for connecting AI applications to external tools and data sources. A compatible server exposes capabilities an assistant can use through a common interface. server notification message handling. Not checked yet Implements LanceDB vector-based tool selection for improved tool routing. Not checked yet Adds MCP router skeleton for future tool routing. Not checked yet Developer MCP extension falls back to .gitignore when no .gooseignore is present. Not checked yet Adds screenshot paste support in the desktop UI. Not checked yet Adds interactive session deletion from the CLI. Not checked yet Adds retries with exponential backoff for the Databricks provider. Not checked yet Supports configurable tool_params_max_length for large tool response handling. Not checked yet New GOOSE_RECIPE_PATH environment variable for discovering recipes in custom paths. Not checked yet GitHub Copilot provider gains streaming support (enabling gpt-4.1 and claude models). Not checked yet Adds Speech MCP extension to the extensions directory. Not checked yet Allows viewing and editing existing recipes in the Desktop UI. Not checked yet CLI hint shown when input is empty (press Enter or Ctrl-J). Not checked yet Google Drive MCP gains label support. Not checked yet Desktop UI supports drag-and-drop file input. Not checked yet Adds recipe_dir configuration for recipe discovery. Not checked yet
27 details in Everything
└──▷ BREAKING ON UPGRADE
! Tool router environment variable names now require a goose prefix (e.g., previously unprefixed vars are renamed).
Goose v1.0.24 adds Venice.ai LLM LLM A large language model, a neural network trained on large amounts of text to recognize patterns and generate language. Applications use it for tasks such as answering, summarizing and writing code. support, tool-loop protection, directory tracking, and recipe parameters
└──▷ GET THIS VERSION
copy
$ git clone --branch v1.0.24 https://github.com/block/goose.git
$ git checkout v1.0.24
└──▷ USE IT
Pin the planner to a specific model and provider without using environment variables
yaml
copy
GOOSE_PLANNER_MODEL: gpt-4o
GOOSE_PLANNER_PROVIDER: openai
Adds Venice.ai as a private open-source LLM provider option Not checked yet Adds tool repetition monitoring to detect and prevent infinite tool-call loops Not checked yet Adds goose run flag to execute sessions with no persistence to disk Not checked yet Adds directory tracking so projects can be resumed by location Not checked yet Adds configurable parameters in recipe files, enabling parameterized automation workflows Not checked yet
+6 more
show less
Adds recipe explanation display in the UI so users can see what a recipe does before running it Not checked yet Supports setting GOOSE_PLANNER_MODEL and GOOSE_PLANNER_PROVIDER in config.yaml Not checked yet Adds token usage progress info alert in the UI Not checked yet Defaults to concise tools mode in the GUI GUI A graphical user interface, controls such as windows, buttons and menus. It lets people use software visually rather than entering commands. for reduced noise Not checked yet Adds providerConfig parameter to exposed goose-llm library functions Not checked yet Auto-generates Kotlin bindings for goose-llm via uniffi-rs proc macros Not checked yet
11 details in Everything
Goose v1.0.23 adds GitHub Copilot provider, a /summarize slash command, session search, and full --resume message loading.
└──▷ GET THIS VERSION
copy
$ git clone --branch v1.0.23 https://github.com/block/goose.git
$ git checkout v1.0.23
└──▷ TRY IT
Trigger an in-session context summarization to free up token space without ending the session.
Adds GitHub Copilot as a supported LLM LLM A large language model, a neural network trained on large amounts of text to recognize patterns and generate language. Applications use it for tasks such as answering, summarizing and writing code. provider. Not checked yet New /summarize slash command lets users trigger context summarization directly from the CLI. Not checked yet Adds a summarize-on-command button for context management in the UI. Not checked yet Loads all previous messages when --resume is passed, enabling full session continuity. Not checked yet Adds search to the sessions list for faster session navigation. Not checked yet
+6 more
show less
Adds a menu item to check Goose Desktop's version on Linux. Not checked yet Expanded ToolCall options for greater tool invocation flexibility. Not checked yet Removes outdated truncation of tool descriptions in OpenAI and Databricks providers, enabling richer tool context. Not checked yet Adds model token limits fallback matching backend logic for more reliable context handling. Not checked yet Adds recipe welcome message support. Not checked yet Allowlist blocks with SSE SSE Server-Sent Events, a browser API and HTTP-based protocol where a server pushes a stream of text events to a client over a single long-lived connection, giving tools a lightweight alternative to WebSockets for one-way real-time updates. downgraded to warning level for less disruptive handling. Not checked yet
11 details in Everything
Goose v1.0.22 adds GitHub-based recipe retrieval, file drag-and-drop, a global hotkey, runtime metrics, and Mac keyboard shortcuts.
└──▷ GET THIS VERSION
copy
$ git clone --branch v1.0.22 https://github.com/block/goose.git
$ git checkout v1.0.22
└──▷ TRY IT
Run a recipe in headless/unattended mode without supplying an interactive prompt, suitable for CI pipelines.
copy
$ goose run --recipe <recipe>.yaml --headless
Run benchmarks and write results to a custom directory for easier artifact collection in CI.
copy
$ goose bench --output-dir <path>
Supports configurable retrieval of recipes via GitHub, enabling teams to share and version-control recipe libraries centrally. Not checked yet Enables drag-and-drop of files directly into the Goose desktop window for faster context attachment. Not checked yet Adds a global hotkey to bring up Goose from anywhere on the desktop. Not checked yet Exposes runtime metrics in completion responses, giving visibility into model execution performance. Not checked yet Adds Mac keyboard shortcuts to the desktop UI. Not checked yet
+6 more
show less
Allows recipes to run in headless mode without requiring a prompt, enabling unattended automation. Not checked yet Adds Azure credential chain logging to aid in diagnosing Azure auth configuration issues. Not checked yet Adds a configurable output directory for benchmark runs. Not checked yet Combines create and reply comment tools into a single unified tool. Not checked yet Stores global and local chat history in localStorage for persistence across sessions. Not checked yet Handles context summarization feedback directly in the UI. Not checked yet
11 details in Everything
Goose v1.0.21 adds context-limit options, session ordering, model-switching without reinit, and spellcheck in the desktop app.
└──▷ GET THIS VERSION
copy
$ git clone --branch v1.0.21 https://github.com/block/goose.git
$ git checkout v1.0.21
Adds ascending/descending ordering support for goose session list. Not checked yet Enables switching or adding a model without reinitializing extensions, reducing disruption mid-session. Not checked yet Presents interactive options to the user when context length context length The maximum amount of text, counted in tokens, a model can take in and produce in one request. Everything the model knows about the task has to fit inside it. is exceeded, with a new context management modal. Not checked yet Adds full spellcheck and autocorrection support for editable content in the desktop app. Not checked yet Adds context limits for the latest Gemini and GPT models. Not checked yet
+7 more
show less
Adds new Google model support in the provider configuration. Not checked yet Adds a goose-llm crate for use by the Goose service. Not checked yet Adds MCP MCP Model Context Protocol, a standard for connecting AI applications to external tools and data sources. A compatible server exposes capabilities an assistant can use through a common interface. router extension disable capability. Not checked yet Reduces the number of Google Drive tools exposed, narrowing the tool surface. Not checked yet Adds bottom bar text truncation with tooltips in the UI. Not checked yet Adds session deletion preview showing the to-be-removed session before confirming. Not checked yet Unifies the permission flow for enabling extensions. Not checked yet
12 details in Everything
Goose v1.0.20 adds session removal, a GUI GUI A graphical user interface, controls such as windows, buttons and menus. It lets people use software visually rather than entering commands. deeplink extension warning, and a bumped tool limit with updated alerts.
└──▷ GET THIS VERSION
copy
$ git clone --branch v1.0.20 https://github.com/block/goose.git
$ git checkout v1.0.20
Adds ability to remove sessions via new remove session capacity. Not checked yet Adds GUI warning when installing an extension from a deeplink. Not checked yet Bumps tool limit and updates associated alerts. Not checked yet
3 details in Everything
Goose v1.0.19 adds o4-mini and GPT-4.1 support, persistent CLI themes, custom OpenAI headers, and an FFI for the Rust library.
└──▷ GET THIS VERSION
copy
$ git clone --branch v1.0.19 https://github.com/block/goose.git
$ git checkout v1.0.19
Adds support for the o4-mini model. Not checked yet Adds per-model prompts for better GPT-4.1 support. Not checked yet Supports OPENAI_CUSTOM_HEADERS being set in the Goose config file. Not checked yet Builds a prototype FFI for the Goose Rust library, enabling external language bindings. Not checked yet GOOSE_CLI_THEME now persists to the config file and applies to future sessions automatically. Not checked yet
+2 more
show less
Adds a permissions page in the bottom menu bar of the chat UI. Not checked yet Adds token and tools alerts in the chat bottom menu bar. Not checked yet
7 details in Everything
Goose v1.0.18 adds tool-level permission control, recipes/custom agents, SSE SSE Server-Sent Events, a browser API and HTTP-based protocol where a server pushes a stream of text events to a client over a single long-lived connection, giving tools a lightweight alternative to WebSockets for one-way real-time updates. extensions, file-based secrets, and GCP Vertex AI support.
└──▷ GET THIS VERSION
copy
$ git clone --branch v1.0.18 https://github.com/block/goose.git
$ git checkout v1.0.18
└──▷ TRY IT
Set model temperature without modifying config — useful for scripting creative or deterministic runs.
copy
$ GOOSE_TEMPERATURE=0.2 goose session
Point Goose at a self-hosted or proxy Anthropic endpoint instead of the default API.
copy
$ ANTHROPIC_HOST=https://my-anthropic-proxy.internal goose session
Adds tool-level permission control in both CLI and UI, letting practitioners approve or deny individual tools before execution. Not checked yet Introduces recipes — custom Goose agent configurations that can be shared via URL deeplinks. Not checked yet Adds a recipe editor UI for building and sharing custom agent configurations. Not checked yet Supports SSE (Server-Sent Events Server-Sent Events An HTTP-based protocol where a server pushes a continuous stream of text events to a client over a single long-lived connection, giving cyber tools a lightweight alternative to WebSockets for one-way real-time data feeds. ) extensions via scheme URI, enabling remote extension connectivity. Not checked yet Adds support for generic GCP Vertex AI Claude and Gemini models as a provider. Not checked yet
+19 more
show less
Adds ANTHROPIC_HOST configuration option for Anthropic in CLI and UI, enabling custom endpoint targeting. Not checked yet Adds a temperature environment variable for controlling model temperature without code changes. Not checked yet Supports file-based secrets for extension configuration. Not checked yet Adds MCP MCP Model Context Protocol, a standard for connecting AI applications to external tools and data sources. A compatible server exposes capabilities an assistant can use through a common interface. router extension discovery and install tool, enabling in-session extension installation. Not checked yet Enables Settings V2 UI with migration logic from Settings V1. Not checked yet Parallelizes extension startup with toast-based error reporting for faster initialization. Not checked yet Adds --with-remote-extension flag for launching Goose with a remote extension. Not checked yet Adds exponential backoff to the Bedrock provider for improved reliability. Not checked yet Supports NO_COLOR environment variable when PrettyPrinter is used. Not checked yet Adds permission field to the list tools API response. Not checked yet Enables frontend tools execution from the UI. Not checked yet Adds full URL extraction support for Google Drive. Not checked yet Auto-initializes goose config.yaml on first run. Not checked yet Enables spellcheck in the chat input. Not checked yet Adds a system theme option to the UI. Not checked yet Adds tooltips throughout the UI. Not checked yet Adds debounce for search and chat text inputs to improve typing responsiveness in large sessions. Not checked yet Accumulates tokens across a session file for accurate total token usage reporting. Not checked yet Adds Google Drive URI verification and example instructions. Not checked yet
24 details in Everything
Goose v1.0.17 adds session sharing, chat search, prompt library, VSCode server support, and input/output token tracking.
└──▷ GET THIS VERSION
copy
$ git clone --branch v1.0.17 https://github.com/block/goose.git
$ git checkout v1.0.17
└──▷ HOW TO FIND IT
Insert a new line mid-prompt without submitting — useful when composing multi-line instructions in the chat UI.
📍 Option + Enter
Adds session sharing from the UI, with pre-configuration via GOOSE_BASE_URL_SHARE environment variable and additional metadata in the shared session view. Not checked yet Adds search functionality to the chat view. Not checked yet Adds a Prompt Library to the UI. Not checked yet Adds the VSCode server to the extensions list. Not checked yet Adds input and output token tracking to SessionMetadata. Not checked yet
+10 more
show less
Adds tool annotations for built-in tools. Not checked yet Adds a named feature flag system for Settings V2. Not checked yet Adds a timeout field to the Settings V2 modal. Not checked yet Adds an additive entry in the Settings V2 model selector. Not checked yet Enables Option+Enter keyboard shortcut to insert a new line in chat input. Not checked yet Adds env-var editor when adding or editing extensions. Not checked yet Adds an allowlist option for the goosed daemon. Not checked yet Adds copy button for extensions and makes extension builtins available at any time. Not checked yet Adds Playwright end-to-end testing setup. Not checked yet Turns the Goose entrypoint into a library function for programmatic use. Not checked yet
15 details in Everything
Goose v1.0.16 adds PDF reading, shareable bots, a /plan command, Databricks+Claude 3.7, custom OpenAI headers, and Java/JDK MCP MCP Model Context Protocol, a standard for connecting AI applications to external tools and data sources. A compatible server exposes capabilities an assistant can use through a common interface. support.
└──▷ GET THIS VERSION
copy
$ git clone --branch v1.0.16 https://github.com/block/goose.git
$ git checkout v1.0.16
└──▷ TRY IT
Generate a structured plan for a complex task before letting Goose execute it — useful for long-horizon or multi-step agentic sessions.
copy
$ goose /plan "Set up a Python project with CI, linting, and tests"
Adds --plan command in CLI to invoke a reasoner with a plan system prompt. Not checked yet Adds PDF reader capability, enabling Goose to ingest PDF files directly. Not checked yet Supports custom HTTP headers for the OpenAI provider. Not checked yet Exports Azure API version as a configurable parameter. Not checked yet Adds Databricks model format with support for Claude 3.7 with extended thinking extended thinking Intermediate reasoning a model writes out before its final answer. Reasoning models generate it automatically and it is often billed as extra output tokens. . Not checked yet
+16 more
show less
Allows setting the OpenAI request timeout from config. Not checked yet Adds retry logic for the Google provider. Not checked yet Adds Java/JDK support for MCP servers running via extensions. Not checked yet Adds a noop tool to the tool shim for testing/passthrough scenarios. Not checked yet Introduces shareable Goose bots for sharing configured assistant profiles. Not checked yet Improves extension security via an environment variable denylist. Not checked yet Supports deep-link extension installs directly from settings. Not checked yet Expands Google Drive integration with comments, replies, folders, shortcuts, and file moves. Not checked yet Adds read and write support for Google Sheets. Not checked yet Adds a description field to the extension configuration and modal. Not checked yet Enables extensions to be turned on automatically at startup. Not checked yet Adds a models dropdown to the UI for switching models inline. Not checked yet Adds a new configure provider flow in the UI. Not checked yet Adds in-app log viewer to the desktop UI. Not checked yet Introduces new toast notifications for system events. Not checked yet Surfaces StdioProcessError messages in the UI for better MCP debugging. Not checked yet
21 details in Everything
Goose v1.0.15 adds Google Drive write/comment support, AWS Bedrock in desktop, and a new CLI workflow builder.
└──▷ GET THIS VERSION
copy
$ git clone --branch v1.0.15 https://github.com/block/goose.git
$ git checkout v1.0.15
Adds Google Drive write tools and comment-read capability, enabling agents to read comments and write files in Drive. Not checked yet Adds image resizing logic for Google Drive with Content::Image support for richer document handling. Not checked yet Adds PKCE PKCE Proof Key for Code Exchange, an OAuth safeguard that binds an authorization code to the client that requested it. An intercepted code cannot be redeemed without the client's matching verifier. -based OAuth2 OAuth2 An open authorization framework (RFC 6749) that lets a service grant limited access to a user's resources on another service without exposing credentials, giving cyber tools a standard way to authenticate against third-party APIs. flow for Google Drive with generic token storage, improving authentication security. Not checked yet Unifies Google Drive read/write scope across all commands so a single token covers all operations. Not checked yet Adds basic AWS Bedrock support to the Goose desktop app as a new model provider. Not checked yet
+3 more
show less
Introduces a CLI workflow build command for constructing and running agent workflows from the command line. Not checked yet Adds a new extensions modal in the UI with auto-start on add and reorganized extensions settings. Not checked yet Adds default metrics for core evaluations to standardize eval measurement. Not checked yet
8 details in Everything
Goose v1.0.14 adds session listing, write-approve mode, Google Sheets, Ollama tool shim, and stdin CLI support.
└──▷ GET THIS VERSION
copy
$ git clone --branch v1.0.14 https://github.com/block/goose.git
$ git checkout v1.0.14
└──▷ TRY IT
List all past Goose sessions to review or resume previous work.
copy
$ goose session list
Pipe a prompt directly into Goose from another command without an interactive session.
copy
$ echo 'Summarize the files in this directory' | goose -i -
Adds session list command to the CLI for viewing past sessions. Not checked yet Adds Google Sheets support in the Google Drive built-in MCP MCP Model Context Protocol, a standard for connecting AI applications to external tools and data sources. A compatible server exposes capabilities an assistant can use through a common interface. server. Not checked yet Adds Ollama tool shim, enabling tool-call support for Ollama-backed models. Not checked yet Supports stdin input in the CLI via -i - or with no arguments for piped/scripted workflows. Not checked yet Moves Google Drive credentials into the system keychain with an optional fallback. Not checked yet
+2 more
show less
Extensions now read from config and load in the background with a visible pending state. Not checked yet Adds a copy-to-clipboard button for error messages in toast notifications. Not checked yet
7 details in Everything
Goose v1.0.13 adds smart-approve mode, memory condensation, PDF/Office file support, a --debug flag, and a full benchmarking framework.
└──▷ GET THIS VERSION
copy
$ git clone --branch v1.0.13 https://github.com/block/goose.git
$ git checkout v1.0.13
└──▷ TRY IT
Capture verbose agent activity to diagnose tool call failures or unexpected model behaviour during a session.
copy
$ goose session --debug
Run a one-shot task with full debug output to trace exactly what the agent does at each step.
copy
$ goose run --debug --instructions 'Summarise the quarterly report' report.pdf
Enables smart-approve mode by default, allowing the agent to automatically approve low-risk tool calls without user interruption. Not checked yet Adds memory condensation for longer-context sessions, keeping conversations coherent beyond normal context limits. Not checked yet Supports reading PDFs natively, plus .doc/.xls and simple HTML files, broadening the range of documents Goose can process. Not checked yet Adds --debug flag to goose session and goose run for verbose diagnostic output during CLI sessions. Not checked yet Implements a tool permission store for persistent tracking of approved/denied tool permissions across sessions. Not checked yet
+12 more
show less
Introduces the Goose Bench framework for functional and regression testing of agent behaviour, including parallel processing in approve mode and eval result writing to an eval directory. Not checked yet Adds an image tool to the developer MCP MCP Model Context Protocol, a standard for connecting AI applications to external tools and data sources. A compatible server exposes capabilities an assistant can use through a common interface. , enabling Goose to capture and process screenshots (including macOS screenshots). Not checked yet Adds Ctrl/Cmd + ↑/↓ keyboard shortcut to navigate message history in the CLI. Not checked yet Supports Databricks OAuth refresh tokens refresh tokens A credential used to obtain a new access token without signing in again. It is sent to the authorization service rather than used directly to call the protected API. via the Databricks API integration. Not checked yet Adds copy message content action to the UI for quick clipboard access to any message. Not checked yet Adds mode completion (tab-complete for modes) in the CLI. Not checked yet Updates config endpoints for use with providers, improving provider configuration flexibility. Not checked yet Enables building Goose inside a Docker container for portable, reproducible deployments. Not checked yet Adds auto-including directories in binary/bench working-dir for the bench framework. Not checked yet Splits required_extensions in bench configuration into builtin and external categories for finer control. Not checked yet Adds UI setting configuration panel for managing settings directly from the desktop interface. Not checked yet Retains session state through view changes in the desktop UI. Not checked yet
17 details in Everything
Goose v1.0.12 adds global prompt templates at startup, form injection for provider modals, and extension management in Settings v2.
└──▷ GET THIS VERSION
copy
$ git clone --branch v1.0.12 https://github.com/block/goose.git
$ git checkout v1.0.12
Loads global prompts at startup using the minijinja templating engine, enabling reusable prompt templates across sessions. Not checked yet Introduces form injection for provider modals in the UI, streamlining provider configuration. Not checked yet Adds extension add and edit functionality in the Settings v2 UI. Not checked yet
3 details in Everything
Goose v1.0.11 adds GCP Vertex AI, Claude 3.7 Sonnet with extended thinking extended thinking Intermediate reasoning a model writes out before its final answer. Reasoning models generate it automatically and it is often billed as extra output tokens. , sessions API, .gooseignore support, and tab-based slash command completion.
└──▷ GET THIS VERSION
copy
$ git clone --branch v1.0.11 https://github.com/block/goose.git
$ git checkout v1.0.11
└──▷ TRY IT
Use tab completion for slash commands and MCP MCP Model Context Protocol, a standard for connecting AI applications to external tools and data sources. A compatible server exposes capabilities an assistant can use through a common interface. prompts in an interactive Goose CLI session.
copy
$ goose
# Inside the session, type '/' then press Tab to browse available slash commands and MCP prompts
Adds GCP Vertex AI as a supported provider platform. Not checked yet Adds Claude 3.7 Sonnet with extended thinking via Anthropic provider. Not checked yet New sessions API enabling viewing and resuming previous sessions. Not checked yet Adds .gooseignore support to restrict Goose's access to specified files or directories. Not checked yet Adds tab-based slash command completion and prompt info completion in the CLI. Not checked yet
+16 more
show less
Adds MCP prompt support via slash commands in the CLI. Not checked yet Implements global command history storage. Not checked yet Supports customizing extension timeout. Not checked yet Supports arbitrary path for sessions in the CLI. Not checked yet Adds goose mode to the chat window with full UI support. Not checked yet Adds context window context window The maximum amount of text, counted in tokens, a model can take in and produce in one request. Everything the model knows about the task has to fit inside it. limit for OpenAI reasoning models. Not checked yet Adds ability for users to turn off smart approve. Not checked yet Builds native Intel Mac app. Not checked yet Adds view and edit support for .goosehints file in the desktop UI. Not checked yet Moves configure goosehints into the MoreMenu for easier access. Not checked yet Adds extensions install link generator. Not checked yet Adds a tutorial built-in extension for the app. Not checked yet Improves CLI default behavior when no command is provided. Not checked yet Installs Python 3.10 in the uvx shim before uv for better compatibility. Not checked yet Keyboard shortcut Command/Ctrl + Comma navigates to Settings view in the desktop UI. Not checked yet Improves the allow-tool UI. Not checked yet
21 details in Everything
Goose v1.0.10 adds goose update, interactive post-run mode, permission prompts before tool calls, and a configurable OpenAI base path.
└──▷ GET THIS VERSION
copy
$ git clone --branch v1.0.10 https://github.com/block/goose.git
$ git checkout v1.0.10
└──▷ TRY IT
Keep your Goose CLI up to date without reinstalling manually.
New goose update command lets users update the Goose CLI to the latest version in place. Not checked yet Interactive mode after run: Goose can drop into an interactive session at the end of a non-interactive run. Not checked yet Permission prompt before each tool call, giving users explicit control over tool execution. Not checked yet Read-only tools now skip the approval prompt automatically, reducing friction for safe operations. Not checked yet Supports setting a custom OpenAI base path, enabling use of OpenAI-compatible endpoints. Not checked yet
+2 more
show less
Settings v2 scaffolding and ConfigProvider integration lay groundwork for a new configuration UI. Not checked yet Alpha providers grid refactor in the UI. Not checked yet
7 details in Everything
Goose v1.0.9 adds an experiment manager for feature flags and consolidates settings config.
└──▷ GET THIS VERSION
copy
$ git clone --branch v1.0.9 https://github.com/block/goose.git
$ git checkout v1.0.9
Adds an experiment manager to control whether individual features are enabled. Not checked yet Consolidates Goose settings into a unified config. Not checked yet Adds a default value for the OpenAI host configuration. Not checked yet
3 details in Everything
Goose v1.0.8 adds a new info command, customizable system prompt templates, OpenAI org/project support, reasoning effort control, and configurable OpenAI host.
└──▷ GET THIS VERSION
copy
$ git clone --branch v1.0.8 https://github.com/block/goose.git
$ git checkout v1.0.8
└──▷ TRY IT
Quickly inspect which config file and directories Goose is using on your machine.
New goose info command displays all directories in use and current configuration at a glance. Not checked yet Adds a fully customizable system prompt template mode, giving users complete control over the system prompt. Not checked yet Supports configuring a custom OpenAI host, enabling use with OpenAI-compatible endpoints. Not checked yet Adds organization and project fields for the OpenAI provider configuration. Not checked yet Supports OpenAI reasoning effort configuration for O1/O3 models. Not checked yet
+2 more
show less
Google Drive search tool gains corpora and pageSize parameters for finer-grained search control. Not checked yet Propagates external_model_message upward on errors, surfacing upstream model error details. Not checked yet
7 details in Everything
Goose v1.0.7 adds Linux computer control support and verbose tool output.
└──▷ GET THIS VERSION
copy
$ git clone --branch v1.0.7 https://github.com/block/goose.git
$ git checkout v1.0.7
Adds Linux computer control capability, enabling automated desktop interaction on Linux systems. Not checked yet Introduces verbose tool output mode for more detailed visibility into tool execution. Not checked yet
2 details in Everything
Goose v1.0.6 adds configurable tool output, image handling, extension management, VertexAI/Claude support, and experimental Windows CLI.
└──▷ GET THIS VERSION
copy
$ git clone --branch v1.0.6 https://github.com/block/goose.git
$ git checkout v1.0.6
Adds goose configure flow for adjusting the amount of tool output shown in sessions. Not checked yet Adds support for handling images mentioned in messages. Not checked yet Enables removing installed extensions via the CLI. Not checked yet Adds server-side endpoints for config management via goose-server. Not checked yet Supports extending the system prompt with custom content. Not checked yet
+6 more
show less
System prompt now includes the current date and response formatting instructions. Not checked yet Simplifies CLI session management. Not checked yet Ports MCP MCP Model Context Protocol, a standard for connecting AI applications to external tools and data sources. A compatible server exposes capabilities an assistant can use through a common interface. prompts into the developer extension. Not checked yet Follows XDG XDG A set of specifications maintained by freedesktop.org that standardise where Linux applications store config, data, and cache files, letting tools respect user-defined paths instead of hardcoding locations. spec on Linux/macOS and uses Windows known folders for config and log paths. Not checked yet Adds experimental Windows support for the Goose CLI. Not checked yet Supports modifying AZURE_OPENAI_API_VERSION via configuration. Not checked yet
11 details in Everything
└──▷ BREAKING ON UPGRADE
! Config and log paths on Linux/macOS now follow the XDG spec; existing setups storing config or logs in non-XDG locations may need to migrate files to the new paths.
Goose v1.0.5 adds AWS Bedrock provider, updated Gemini models, multi-value CLI extension options, and visible session aliases.
└──▷ GET THIS VERSION
copy
$ git clone --branch v1.0.5 https://github.com/block/goose.git
$ git checkout v1.0.5
Adds AWS Bedrock as a new LLM LLM A large language model, a neural network trained on large amounts of text to recognize patterns and generate language. Applications use it for tasks such as answering, summarizing and writing code. provider. Not checked yet Updates available Gemini models. Not checked yet Supports multiple values in a single CLI option when adding extensions. Not checked yet Makes the CLI session command alias visible in help output. Not checked yet Replaces the prompt for Ollama smaller models for improved compatibility. Not checked yet
+2 more
show less
Respects the terminal emulator's base colors for better theme integration. Not checked yet Adds GitHub project links on extension cards in the UI. Not checked yet
7 details in Everything
Goose v1.0.4 adds Azure OpenAI as a provider and configures Electron to open links in new windows.
└──▷ GET THIS VERSION
copy
$ git clone --branch v1.0.4 https://github.com/block/goose.git
$ git checkout v1.0.4
Adds Azure OpenAI as a supported LLM LLM A large language model, a neural network trained on large amounts of text to recognize patterns and generate language. Applications use it for tasks such as answering, summarizing and writing code. provider. Not checked yet Configures Electron app to open external links in a new window instead of in-app. Not checked yet
2 details in Everything
Goose v1.0.3 adds a CONFIGURE=false install option and Ollama host configuration support.
└──▷ GET THIS VERSION
copy
$ git clone --branch v1.0.3 https://github.com/block/goose.git
$ git checkout v1.0.3
└──▷ TRY IT
Install Goose non-interactively in CI or automated environments without being prompted to configure a provider.
copy
$ CONFIGURE=false curl -fsSL https://github.com/block/goose/releases/latest/download/install.sh | bash
Adds CONFIGURE=false option to the install script to skip interactive configuration during automated or scripted installs. Not checked yet Supports setting a custom Ollama host, enabling Goose to connect to remote or non-default Ollama instances. Not checked yet Updates the UI to expose Ollama host configuration directly in the interface. Not checked yet
3 details in Everything
Goose v0.9.9 adds a browser toolkit, tracing support, a license checker, and defaults to the synopsis toolkit.
└──▷ GET THIS VERSION
copy
$ git clone --branch v0.9.9 https://github.com/block/goose.git
$ git checkout v0.9.9
Adds a browser toolkit, enabling Goose to interact with web browsers as a tool. Not checked yet Adds tracing option to the run command, grouping traces under the session name for observability. Not checked yet Adds a license checker for auditing dependency licenses. Not checked yet Defaults Goose to the synopsis toolkit out of the box. Not checked yet Adds support for optional parameters in JSON schema conversion in exchange, broadening tool-call compatibility. Not checked yet
+2 more
show less
Adds a list moderators capability. Not checked yet Includes the new Anthropic model in recommended config. Not checked yet
7 details in Everything
Goose v0.9.6 adds web browsing, session resumption, and a synopsis-based core loop.
└──▷ GET THIS VERSION
copy
$ git clone --branch v0.9.6 https://github.com/block/goose.git
$ git checkout v0.9.6
Adds web browsing capability to the core Goose loop. Not checked yet Enables resuming a previous session with a dedicated run flag. Not checked yet Introduces a synopsis approach for the core Goose loop to improve context handling. Not checked yet
3 details in Everything
Goose v0.9.5 adds Google vision support and updates the default Ollama model to qwen2.5.
└──▷ GET THIS VERSION
copy
$ git clone --branch v0.9.5 https://github.com/block/goose.git
$ git checkout v0.9.5
Adds vision support for Google provider, enabling image-based interactions. Not checked yet Updates the default Ollama model from mistral-nemo to qwen2.5. Not checked yet
2 details in Everything
Goose v0.9.4 adds Groq and o1 model support, a new run command, provider listing, keychain API key storage, and global goosehints.
└──▷ GET THIS VERSION
copy
$ git clone --branch v0.9.4 https://github.com/block/goose.git
$ git checkout v0.9.4
└──▷ TRY IT
Check which LLM LLM A large language model, a neural network trained on large amounts of text to recognize patterns and generate language. Applications use it for tasks such as answering, summarizing and writing code. providers are configured and available before starting a session.
copy
$ goose providers list
Adds Groq as a supported LLM provider. Not checked yet Supports OpenAI o1-preview and o1-mini deep thinking reasoner models. Not checked yet New goose run command for non-interactive session execution. Not checked yet New goose providers list command to enumerate available providers. Not checked yet Saves API keys to the system keychain automatically for users. Not checked yet
+3 more
show less
Adds session management guards to protect active sessions. Not checked yet Adds support for a global user-level goosehints file for persistent instructions. Not checked yet Adds local Langfuse tracing option for observability. Not checked yet
8 details in Everything
Goose v0.9.3 adds shell completions, Jira toolkit, sandbox mode, markdown plans, and session auto-save.
└──▷ GET THIS VERSION
copy
$ git clone --branch v0.9.3 https://github.com/block/goose.git
$ git checkout v0.9.3
└──▷ TRY IT
Generate and install shell completions so you can tab-complete Goose subcommands and flags.
copy
$ goose shell-completions
Adds shell-completions subcommand for generating shell tab-completion scripts. Not checked yet Introduces a Jira toolkit for interacting with Jira from within Goose sessions. Not checked yet Supports running Goose inside a Docker-style sandbox for isolated execution. Not checked yet Enables support for markdown-formatted plans. Not checked yet Auto-saves sessions before the next user input to preserve context. Not checked yet
+3 more
show less
Tracks cost and token usage in the log file for session auditing. Not checked yet Adds --version option to display the current Goose version. Not checked yet Gives commands the ability to execute logic, expanding extensibility of the command system. Not checked yet
8 details in Everything
Goose v0.9.0 adds Jinja-templated goosehints, per-repo hints, toolkit listing, and regex-based dangerous command detection.
└──▷ GET THIS VERSION
copy
$ git clone --branch v0.9.0 https://github.com/block/goose.git
$ git checkout v0.9.0
Supports Jinja templating in goosehints files for dynamic hint configuration. Not checked yet Adds per-repository hints so projects can ship their own Goose context alongside code. Not checked yet New command to show all available toolkits at a glance. Not checked yet Adds regex-based checks to detect and flag potentially dangerous shell commands. Not checked yet
4 details in Everything