The daily firehose — everything the toolchain shipped today, already filtered.
We read every release from the 1309 tools on our watchlist at the source — GitHub and GitLab release notes, vendor release pages and changelogs, project blogs and feeds, vendor press releases, and the source code behind the tag. Bug-fix-only releases and non-product newsroom noise are dropped; what's left is summarized down to the new capability, how to try it, and any screenshots or videos the release itself published. Every entry links to the sources it was built from.
READ AS
Focus:What features shipped and what problems they solve, plus launches, packaging and where the market is moving.
Leaves out:Command-line flags, API-only deltas, usage examples, and releases that are only refinements.
Focus:New security controls, new APIs, and product improvements you can put to work, plus anything an upgrade breaks.
Leaves out:Launch and marketing news, governance, pricing, and documentation changes with no operational effect.
Focus:Dashboards, reporting, and anything you can quantify: metrics, coverage, risk scores, compliance impact, vendor moves.
Leaves out:Settings, flags, API and command-line detail, library bumps, and minor releases with no reporting or posture angle.
Focus:API, SDK and MCP changes, new libraries, breaking changes in what you import, and cyber tooling.
Leaves out:Governance, SOC operations, and vendor launches with no developer surface.
AMOUNT
DETAIL
ORDER
This issue came out before editions were ranked, so it shows the Standard edition in issue order.
Cloudsmith is a cloud-native, universal package management platform that enables teams to securely host, manage, and distribute software artifacts across multiple formats.
Cloudsmith CLI gains automatic OIDCOIDCOpenID Connect, a sign-in protocol built on OAuth 2.0. An application uses identity information from a trusted provider instead of maintaining a separate password for each user. auth and Docker credential discovery; cooldown policies expand to Maven packages.
Introduces a Docker credential helper in the CLI to automatically authenticate to Cloudsmith registries.
Extends cooldown policies to Maven packages, enabling supply-chain protection against newly published versions that haven't undergone sufficient scrutiny.
trufflehog discovers, classifies, and validates leaked credentials
TruffleHog v3.96.0 adds HashiCorp Vault token, Duo API Secret Key, and SonarQube Cloud scoped token detectors plus updated Cloudflare credential format support.
└──▷ GET THIS VERSION
$ git clone --branch v3.96.0 https://github.com/trufflesecurity/trufflehog.git
# already have the repo? check out this version:$ git checkout v3.96.0
Adds HashiCorp Vault token detector supporting service tokens, periodic tokens, and legacy tokens with live verification against Vault's auth/token/lookup-self endpoint.
Adds Duo API Secret Key detector for scanning Duo credentials.
ALEAPP v2026.2.0 adds C2PA AI provenanceprovenanceEvidence 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. parsing, Mister Skinnylegs browser artifact integration, and enhanced Samsung clipboard/CMH extraction.
└──▷ GET THIS VERSION
$ git clone --branch v2026.2.0 https://github.com/abrignoni/ALEAPP.git
# already have the repo? check out this version:$ git checkout v2026.2.0
New C2PA Content Provenance artifact parses AI-generated media provenance data including signer certificate, signing time, case-insensitive media path matching, and a Media State column grouping manifest rows per media file.
New Mister Skinnylegs artifact plugin integrates browser artifact processing targeting every Chromium-based browser or browser-style app.
Samsung Honeyboard clipboard artifact reworked with WALWALWrite-Ahead Log: a durability mechanism in databases where changes are recorded to an append-only log before being applied to main storage, giving tools crash recovery and atomic transactions without full-page rewrites. recovery, SEFT support, and dual-schema handling (including old-schema UID resolution from ABX packages.xml on Android 12+).
Samsung CMH now parsed directly from the files table, recovering records on Android 15 devices where the images view was dropped.
Finite State Platform provides software supply chain security and firmware analysis tools to identify vulnerabilities and risks in IoT and embedded devices.
Finite State adds match evidence to findings CSV exports, carry-forward component exclusions, and modernized Bamboo/Jenkins CI/CD plugins.
└──▷ HOW TO FIND IT
After running a findings export, use the match evidence columns to identify exactly which files contain a vulnerable component and prioritize remediation.
📍In the console, navigate to a version's Findings tab, click Export CSV, and open the downloaded file — each row now includes file path columns under match evidence alongside the existing vulnerability fields.
Adds match evidence (file paths associated with each vulnerable component) to the findings CSV export, enabling precise remediation targeting.
Carries component exclusions forward to newly scanned versions by default, reducing repeated manual configuration.
Updates Bamboo and Jenkins CI/CD plugins to use the upgraded API, with the Jenkins plugin now published to the Jenkins Update Center.
Queues binary scan uploads that exceed the concurrency limit instead of rejecting them, and automatically cleans up scans stuck in a started state.
Recomputes policy violations and warnings automatically on very large versions when policy settings or vulnerability data change, without requiring a new scan.
+3 moreshow less
Persists table sorting across your session on the Bill of MaterialsBill of MaterialsA structured list of every component, library, and dependency that makes up a piece of software or hardware, used by cyber tools to track what is inside a product and surface known vulnerabilities in its parts., All Components, and Project Dependencies tables.
Adds column tooltips, consistent number formatting, user display names, and clickable policy-status badges to data tables across the platform.
Accepts legacy identifiers across the upgraded API, including the legacy projectVersion query parameter as a deprecated alias for versionId.
Tines is a no-code automation platform that helps security and operations teams build workflows to automate repetitive tasks and incident response processes.
Tines AI Agent actions and Workbench can now search the web, powered by Parallel.ai.
ProjectDiscovery Cloud is a platform providing cloud-based security scanning and reconnaissance tools for identifying vulnerabilities and assets in applications.
ProjectDiscovery Cloud adds Internal Network Scanning to expose reachable internal assets the way an attacker inside the network would.
New Internal Network Scanning capability scans assets from inside the network to surface reachable exposures that external scanners miss.
ConductorOne is an identity governance platform that automates access management, provisioning, and compliance across enterprise applications and infrastructure.
ConductorOne adds AI agent policy steps, access-graph search, group/role reviews, MCPMCPModel 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. per-user auth, and scoped API keys.
└──▷ HOW TO FIND IT
Set up an AI agent step in a policy to auto-evaluate low-risk access requests and reassign to super admins if it can't decide.
📍In the console, go to Policies › [your policy] › Edit, add a new step, select Agent as the step type, set Agent mode to Full control, set Agent failure action to Reassign to super admins, and enter Agent instructions such as 'Approve if the requester is in the contractors group and the resource is tagged low-risk; otherwise deny.'
Scope a personal API key to read-only to safely hand credentials to a reporting script without granting full account access.
📍In the console, go to your profile › API Keys › Create API Key, then open the Limit Scopes selector and choose the read-only role before saving.
Include groups and roles in a quarterly access review campaign to catch over-provisioned shared identities alongside user grants.
📍In the console, go to Access Reviews › Campaigns › Create Campaign, open the Scope tab, and set the Grant types selector to Users and resources before launching.
Adds AI Agent steps to request and review policies — configure agent mode (comment-only, full control, or policy-step changes), failure action (reassign or skip), and free-text approval instructions; evaluations appear in audit log and AI > Conversations.
Extends request catalog search to surface entitlements that indirectly grant access to a searched resource, with a View access graph link showing the full grant chain.
Access review campaigns can now scope to groups and roles (not just users) via a new Grant types selector with Users only, Resources only, or Users and resources options.
Campaign templates support custom recurrence intervals (e.g. every 2 days or every 3 months) and configurable campaign run duration.
IAMIAMIdentity and Access Management — a framework of policies and services for controlling who can authenticate to systems and what resources they are authorized to use, letting cyber tools enforce least-privilege and audit access. policy documents on AWS resources now render as a readable, collapsible JSON tree via a View policy button in resource-level access reviews and task details.
+12 moreshow less
Requests for entitlements created ahead of connector sync (e.g. via Terraform) now queue and auto-resume after sync, instead of failing immediately.
C1 groups can now be added to a request catalog or access profile, making group membership requestable through standard approval, duration, and revocation flows.
Adds Retry failed provisioning action on access tasks, allowing provisioning retries without canceling or losing collected approvals.
Step-up (MFAMFAMulti-factor authentication, requiring proof from different categories, such as a password and a hardware key, to sign in. It reduces reliance on a password alone.) approvals that also require a comment can now be completed together in a single flow.
Per-user authentication for external MCP servers now supports OAuth2OAuth2An 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., bearer tokenbearer tokenA 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., custom header, or basic auth — replacing shared admin credentials.
New Data lineage view for AI-generated reports shows exact queries behind each number and lets users copy lineage for audit workpapers.
C1 AI now renders multiple requested charts as an aligned grid and can produce composed reports combining stat cards, trend charts, and data tables in one response.
New Generate action in the CELCELCommon Expression Language, an open-source expression language maintained by Google, used to embed lightweight, safe policy evaluation logic into tools without running a full scripting engine. editor converts plain-English descriptions directly into CEL expressions without requiring iterative AI chat.
Tool enablement (require enablement vs. auto-enable) can now be set per MCP server, independent of the tenant-wide AI governance default.
Admins can now register MCP servers that fail relaxable OAuth discovery checks (e.g. issuer mismatch, cross-host authorization server) by reviewing and acknowledging flagged findings.
Hosted MCP servers without an admin-set tool prefix now default to the prefix declared by the connector itself, keeping tool names recognizable to AI clients.
Any user can now use the Limit Scopes selector when creating a personal API key to restrict it to a narrower role (e.g. read-only), without requiring Super Admin access.
SailPoint is an identity security platform that helps organizations manage and govern user access to applications, data, and systems across the enterprise.
SailPoint ISC gets a centered navigation bar for admins, end users, and Data Access Security.
Moves the Admin tab and its megamenu to the top center of the screen for a symmetrical, at-a-glance view of all menu and submenu items.
Relocates the + Add/Create quick-access button to the center of the main menu bar, making it reachable from anywhere on the page.
Centers core navigation items (home icon, primary product tabs) in the header for end users to reduce eye and mouse fatigue.
Applies the same centered navigation layout to Data Access Security, creating a consistent visual experience when switching between ISC and DASDASDirect-Attached Storage: disks connected directly to a host machine rather than over a network, giving tools low-latency, high-throughput access to raw storage without SAN or NAS overhead..
SailPoint ISC adds custom Forms for Access Request, collecting contextual data from requesters before submission.
Enables admins to define custom forms and attach them to Roles, Access Profiles, or Entitlements, prompting requesters for supplementary data before submitting an access request.
Passes submitted form data to approvers as name-value pairs, giving decision-makers full context at approval time.
Makes form data accessible throughout the request flow, including before-provisioning rules, enabling custom provisioning logic driven by requester-supplied data.
Automatically copies form data across multiple form instances when a requester asks for several items that each require the same form.
Supports external access request portals (e.g., SailPoint ServiceNow Catalog Integration) by allowing supplementary form data to be passed into the request flow independent of the native UI.
CISO Assistant is a one-stop-shop GRC platform for Risk Management, AppSec, Compliance & Audit, TPRM, BIA, Privacy, and Reporting. It supports 150+ global frameworks with automatic control mapping, including ISO 27001, NIST CSF, SOC 2, CIS, PCI DSS, NIS2, DORA, GDPR, HIPAA, CMMC, and more.
CISO Assistant v3.20.3 adds TISAXTISAXA vehicle-industry information-security assessment standard, maintained by the ENX Association, that lets automotive suppliers prove their data-handling practices to OEMs without each OEM running its own separate audit. v2027, ENISAENISAThe European Union Agency for Cybersecurity, an EU body that publishes threat landscapes, guidelines, and standards to help member states and organisations raise their cyber defences. Cyber Resiliency frameworks, a Power BI native connector, Slovenian UI, and custom fields on security exceptions.
└──▷ GET THIS VERSION
$ git clone --branch v3.20.3 https://github.com/intuitem/ciso-assistant-community.git
# already have the repo? check out this version:$ git checkout v3.20.3
Adds TISAX v2027 framework library for automotive information security assessments.
Kiro is an AI-powered IDE by AWS that uses "specs" to translate requirements into structured plans, then automatically generates and implements code.
Claude Opus 5 is now available in Kiro IDE, CLI, and Web with a 1M context windowcontext windowThe 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. and improved agentic coding performance.
└──▷ HOW TO FIND IT
Switch to Claude Opus 5 after restarting your IDE to tackle large multi-file refactors or end-to-end feature work.
📍Restart the Kiro IDE or CLI, then open the model selector and choose Claude Opus 5.
Adds Claude Opus 5 to the Kiro IDE, CLI, and Web model selector with a 1M token context window.
Supports Claude Opus 5 for multi-file features, large refactors, and end-to-end feature work with self-verification and iteration.
Enables Claude Opus 5 multi-agent coordination with fewer conflicts between parallel agents.
Rolls out to Kiro Pro, Pro+, Pro Max, and Power customers in us-east-1 and eu-central-1 with cross-region inference at a 2.2x credit multiplier.
Pi v0.82.1 adds Claude Opus 5 support, bearer auth for Anthropic-compatible gateways, and faster model catalog revalidation.
└──▷ GET THIS VERSION
$ git clone --branch v0.82.1 https://github.com/earendil-works/pi.git
# already have the repo? check out this version:$ git checkout v0.82.1
└──▷ TRY IT
Authenticate against an Anthropic-compatible gateway that requires bearer tokenbearer tokenA 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 instead of an API key.
$ export ANTHROPIC_AUTH_TOKEN=<your-bearer-token>
Adds Claude Opus 5 model support on Anthropic and Amazon Bedrock, including adaptive thinking (with xhigh level), inference profiles, and prompt caching.
Adds ANTHROPIC_AUTH_TOKEN bearer authentication for Anthropic-compatible gateways that require Authorization: Bearer headers, covering compaction and branch summaries.
Speeds up pi.dev model catalog refreshes with If-None-Match conditional requests, so unchanged provider catalogs return a 304 instead of a full download.
Exposes the outputPad setting to custom message renderers via the Extensions API.
Lightweight coding agent that runs in your terminal
Adds configurable code-mode host fallback control and code-mode tool names in Responses Lite metadata.
└──▷ GET THIS VERSION
$ git clone --branch rust-v0.146.0-alpha.10 https://github.com/openai/codex.git
# already have the repo? check out this version:$ git checkout rust-v0.146.0-alpha.10
└──▷ USE IT
Prevent silent fallback to embedded V8 so failures in the standalone code-mode host surface as tool output rather than being swallowed.
toml
# In your Codex config file:
[features.code_mode_host]
disable_in_process_fallback = true
Adds disable_in_process_fallback option to features.code_mode_host config table, letting operators prevent silent fallback to embedded V8 when the standalone code-mode host fails to start.
Adds code_mode_tool_names field to Responses Lite turn metadata, exposing normalized code-mode identifier-to-ToolName mappings for each turn.
Scopes pending network approvals to a turn and execution, coalesces duplicate requests within one execution, and assigns unique per-execution identifiers to user approval requests and permission hooks.
3 details in Everything
0 more releases in this issue2 more releases in this issue · 2026-07-24 → 2026-07-25
Codex CLI adds ent26 enterprise plan support, MCPMCPModel 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. proxy-aware auth discovery, ephemeral thread forks, and workspace plugin publish capability.
└──▷ GET THIS VERSION
$ git clone --branch rust-v0.146.0-alpha.8 https://github.com/openai/codex.git
# already have the repo? check out this version:$ git checkout rust-v0.146.0-alpha.8
Supports the ent26 enterprise plan tier, including authentication, rate-limit payloads, cloud-config eligibility, and usage-limit guidance.
Routes MCP OAuthMCP OAuthAn authentication layer for the Model Context Protocol that lets MCP clients obtain delegated access tokens so cyber tools can call MCP servers on a user's behalf without handling raw credentials. discovery and authentication status through each server's configured runtime HTTP client, enabling reliable auth for MCP servers behind proxies.
Supports ephemeral forks of paginated threads via thread/fork with excludeTurns: true, preserving selected history and fork preview without creating a persistent rollout.
Exposes canPublishToWorkspace capability metadata in plugin share contexts and plugin/share/save responses so clients can offer workspace-directory publishing.
Codex CLI adds Agent Plugins manifest support, background MCPMCPModel 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. prewarming, and paginated thread forks.
└──▷ GET THIS VERSION
$ git clone --branch rust-v0.146.0-alpha.7 https://github.com/openai/codex.git
# already have the repo? check out this version:$ git checkout rust-v0.146.0-alpha.7
Adds support for Agent Plugins 1.0 manifests: recognizes root plugin.json files and maps their metadata, skills/, and mcp.json into Codex plugin manifests.
Applies Codex-specific apps, hooks, and interface settings from the inline com.openai extension in Agent Plugins manifests, with .codex-plugin/plugin.json as a fallback overlay.
Adds background prewarming of MCP runtime updates so model steps are less likely to stall waiting for MCP rebuilds after config or auth changes.
Refreshes MCP runtimes automatically when session authentication changes (auth mode, account details, or token).
Reconnects all configured MCP servers when an explicit refresh is triggered, rather than reusing existing connections.
+4 moreshow less
Exposes executor skills through skill tools and enables resource reads for explicit executor skills.
Supports paginated thread forks.
Preserves output from hooks that exit before reading stdin (fast-exiting hooks no longer lose stdout).
Allows hosts to customize the model-visible wait_for_environment tool and environment_id descriptions via WaitForEnvironmentToolConfig.
Crush v0.87.0 adds MCP OAuthMCP OAuthAn authentication layer for the Model Context Protocol that lets MCP clients obtain delegated access tokens so cyber tools can call MCP servers on a user's behalf without handling raw credentials. 2.1, Claude Channels MCPMCPModel 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, and a system-wide config at /etc/crush/crush.json
└──▷ GET THIS VERSION
$ git clone --branch v0.87.0 https://github.com/charmbracelet/crush.git
# already have the repo? check out this version:$ git checkout v0.87.0
Adds OAuth 2.1OAuth 2.1A consolidated draft of the OAuth authorization framework that merges OAuth 2.0 with its most important security BCP updates, dropping unsafe flows so tools can delegate user access without handling credentials directly. authorization support for HTTP and SSESSEServer-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. MCP servers, unblocking MCPs that require authentication.
Supports the Claude Channels MCP extension, allowing MCP servers to push real-time notifications into Crush.
Supports system-wide configuration via /etc/crush/crush.json, improving NixOS and multi-user deployments.
Significant LSPLSPLanguage Server Protocol, the standard way editors get code intelligence such as completions, go-to-definition and errors from a separate per-language server. Coding agents use it to understand code. performance improvement: filters servers before searching $PATH, reducing CPUCPUThe main general-purpose processor in a computer. It runs application instructions and coordinates work with memory and other hardware. and memory usage in large repositories.
Autonomous coding agent as an SDK, IDE extension, or CLI assistant.
Cline v4.0.11 adds Claude Opus 5 across six providers and Moonshot Kimi K3 support.
└──▷ GET THIS VERSION
$ git clone --branch v4.0.11 https://github.com/cline/cline.git
# already have the repo? check out this version:$ git checkout v4.0.11
Adds Claude Opus 5 support across Anthropic, Claude Code, Bedrock, Vertex, Cline, and OpenRouterOpenRouterA 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. providers, including 1M context windowcontext windowThe 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. variants.
Adds Moonshot Kimi K3 model support.
Includes host plugin version in all telemetry events.
Claude Code is an agentic coding tool that lives in your terminal, understands your codebase, and helps you code faster by executing routine tasks, explaining complex code, and handling git workflows - all through natural language commands.
Claude Opus 5 lands as the new default Opus model, with 1M context, strict sandbox networking, and depth-3 nested subagents.
└──▷ GET THIS VERSION
$ git clone --branch v2.1.219 https://github.com/anthropics/claude-code.git
# already have the repo? check out this version:$ git checkout v2.1.219
└──▷ USE IT
Lock sandbox network access to only approved hosts so any unexpected outbound call is silently denied without an interactive prompt.
json
# In your Claude Code settings file (e.g. ~/.claude/settings.json)
{
"sandbox": {
"network": {
"strictAllowlist": true
}
}
}
Cap the depth of nested subagent spawning at 1 (flat) to reproduce pre-v2.1.219 behaviour in automation pipelines that shouldn't recurse.
$ CLAUDE_CODE_MAX_SUBAGENT_SPAWN_DEPTH=1 claude -p "refactor the auth module"
Pin the Dynamic workflow size guideline org-wide via settings so all users get it without having to touch /config.
json
# In a shared settings file
{
"workflowSizeGuideline": "medium"
}
Adds Claude Opus 5 (claude-opus-5) as the default Opus model — 1M context windowcontext windowThe 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., fast mode at $10/$50 per Mtok.
Adds sandbox.network.strictAllowlist setting to silently deny any sandboxed-command network request to a non-allowlisted host (no prompt).
Adds DirectoryAdded hook that fires whenever /add-dir or the SDK register_repo_root control request registers a new working directory mid-session.
Adds mcp_server_errors field to the headless stream-json init event, listing --mcp-config entries skipped by config validation; terminal runs surface a startup warning.
Adds workflowSizeGuideline settings key so the Dynamic workflow size advisory can be set from any settings file (hides the /config row while active).
+4 moreshow less
Adds nested subagent forwarding in stream-json: depth-2+ subagents now appear in the stream when --forward-subagent-text is set, keyed by their spawning Agent tool_use id.
Subagents can now spawn nested subagents up to depth 3 (was 1); set CLAUDE_CODE_MAX_SUBAGENT_SPAWN_DEPTH=1 to revert to flat spawning.
Adds current default workflow size to the running-workflow status line with a pointer to /config for changes.
Removes Opus 4.7 from fast mode; /fast now applies to Opus 5 and Opus 4.8 only.
9 details in Everything
└──▷ BREAKING ON UPGRADE
!Opus 4.7 is removed from fast mode; /fast now applies only to Opus 5 and Opus 4.8.
!Dynamic workflows now default to a medium size guideline (fewer than 15 agents); workflows relying on the previous unrestricted default will be subject to this new cap unless overridden in /config.
!Managed MCPMCPModel 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. allowlist/denylist ${VAR} entries now resolve from the startup environment and managed-settings env instead of the settings-file env, which may change which values are substituted.
Daytona is a self-hosted development environment manager that automates provisioning, configuration, and management of standardized, secure dev environments across any infrastructure.
Daytona adds sandbox metrics visibility and configurable auto-pause intervals for sandboxes.
Adds sandbox metrics to surface resource and usage data for running sandboxes.
Supports configurable auto-pause intervals, letting users control how long a sandbox idles before pausing.
Daytona SDKs gain WebSocketWebSocketA 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.-based sandbox event subscriptions for real-time sandbox lifecycle updates.
Adds WebSocket sandbox event subscription support to the Daytona SDKs, enabling real-time event-driven monitoring of sandbox state changes.