The Cyber Toolchain — issue 001, June 29, 2026

Snyk launches Agentic Development Security (ADSADSAlternate Data Streams: an NTFS feature that lets a file carry hidden secondary data blobs under the same path, giving cyber tools a place to stash metadata or detect covert storage by malware.) to govern AI agent supply chains, behavior, and generated code in real time.

5 details in Everything

Snyk offers full AI Security Platform free to open source maintainers, plus a new Remediation Agent for merge-ready fixes.

2 details in Everything

AFL++ v5.01c adds adaptive MOpt, QEMUQEMUAn open-source machine emulator and virtualizer that lets software run a guest OS or CPU architecture on a different host, giving cyber tools an isolated, inspectable execution environment without dedicated hardware. non-colliding coverage, a new qemu_bridge mode, and faster macOS persistent fuzzing.

$ git clone --branch v5.01c https://github.com/AFLplusplus/AFLplusplus.git
# already have the repo? check out this version:
$ git checkout v5.01c

$ afl-fuzz -i seeds_dir -o output_dir -Q -- /path/to/target @@

$ AFL_USE_UBSAN=1 afl-cc -o target_ubsan target.c && afl-fuzz -i seeds_dir -o output_dir -- ./target_ubsan @@

$ AFL_FAST_CHILD_SYNC=0 afl-fuzz -i seeds_dir -o output_dir -- ./target @@

6 details in Everything

AFL++ v5.00c adds AGPL license, Ball-Larus path coverage, a 6-mode bug-detection pass, new cmplog scheduling, and macOS/futex speed gains.

$ git clone --branch v5.00c https://github.com/AFLplusplus/AFLplusplus.git
# already have the repo? check out this version:
$ git checkout v5.00c

$ AFL_LLVM_PATH=3 afl-clang-lto -o target_instrumented target.c -lz

$ AFL_LLVM_BUG=1 afl-clang-fast -o target_bug target.c && afl-fuzz -i seeds/ -o out/ -l M -- ./target_bug @@

$ AFL_LLVM_BUG_ALLOCSIZE_FUNCS=my_alloc AFL_LLVM_BUG_ALLOCSIZE_FREE_FUNCS=my_free AFL_LLVM_BUG_ALLOCSIZE_DERIVE=1 afl-clang-fast -o target_alloc target.c && afl-fuzz -i seeds/ -o out/ -l Z -- ./target_alloc @@

14 details in Everything

TruffleHog v3.95.6 adds six new secret detectors (Datadog API key, Tly, Wit, Rev, User, Enigma) and lifts the 64 KB line-scanning cap to 10 MB.

$ git clone --branch v3.95.6 https://github.com/trufflesecurity/trufflehog.git
# already have the repo? check out this version:
$ git checkout v3.95.6

$ trufflehog git https://github.com/<org>/<repo> --results=verified

9 details in Everything

TruffleHog v3.95.5 adds GitLab OAuth, SpectralOps, and AWS AppSync detectors plus ASPX/entity-encoded HTML decoding.

$ git clone --branch v3.95.5 https://github.com/trufflesecurity/trufflehog.git
# already have the repo? check out this version:
$ git checkout v3.95.5

$ trufflehog git https://github.com/<org>/<repo> --results=verified --json

7 details in Everything

TruffleHog v3.95.4 adds GitLab OAuth, AWS AppSync, and SpectralOps detectors plus richer database ExtraData and expanded HTML decoding.

$ git clone --branch v3.95.4 https://github.com/trufflesecurity/trufflehog.git
# already have the repo? check out this version:
$ git checkout v3.95.4

$ trufflehog git https://gitlab.com/myorg/myrepo --results=verified --json

$ trufflehog filesystem path/to/webapp --results=verified,unknown --json

8 details in Everything

Cloudflare opens self-managed OAuth to all customers, enabling any developer to build scoped, delegated integrations on the Cloudflare API.

⤢

5 details in Everything

Cloudflare introduces Temporary Accounts for AI agents, enabling instant Worker deployments via wrangler deploy --temporary with no sign-up required.

⤢ ⤢ ⤢ ⤢

$ wrangler deploy --temporary

5 details in Everything

Cloudflare Agents SDK gains durable execution, sandboxed code execution, a virtual filesystem, and dynamic workflows for production agent harnessesagent harnessesThe software around an AI agent that manages its tools, execution loop, state and limits. It determines how the model's proposed actions are run and checked..

⤢ ⤢ ⤢

import { Agent } from "agents";
import type { FiberRecoveryContext } from "agents";

class MyAgent extends Agent {
  async doWork() {
    await this.runFiber("my-task", async (ctx) => {
      const step1 = await expensiveOperation();
      ctx.stash({ step1 });
      const step2 = await anotherExpensiveOperation(step1);
      this.setState({ ...this.state, result: step2 });
    });
  }

  async onFiberRecovered(ctx: FiberRecoveryContext) {
    if (ctx.name !== "my-task") return;
    const { step1 } = (ctx.snapshot ?? {}) as { step1?: unknown };
    if (step1) {
      const step2 = await anotherExpensiveOperation(step1);
      this.setState({ ...this.state, result: step2 });
    }
  }
}

$ flue add channel slack

7 details in Everything

Cloudflare One Stack delivers agent-ready skill files for Zero Trust deployment, migration, and troubleshooting automation.

7 details in Everything

Cloudflare DMARC Management reaches GA with IP-level threat investigation, unified auth record analysis, and SPFSPFAn email authentication standard that lets a domain owner publish, via DNS, which mail servers are authorized to send on its behalf, so receiving servers can reject spoofed senders. lookup auditing.

⤢ ⤢ ⤢ ⤢

4 details in Everything

Cloudflare Security Insights now scans all accounts automatically, with frequency doubled across all plans.

⤢ ⤢ ⤢ ⤢

3 details in Everything

Cloudflare launches Application Services for Private Origins, routing public traffic to private IPs via WAFWAFA reverse proxy that inspects HTTP/HTTPS traffic and blocks requests matching known attack patterns such as SQL injection or XSS, sitting between clients and a web application to filter malicious input before it reaches the server., bot mgmt, Workers, and Spectrum.

⤢ ⤢

$ curl -X POST 'https://api.cloudflare.com/client/v4/zones/{zone_id}/dns_records' \
  -H 'Authorization: Bearer <token>' \
  -H 'Content-Type: application/json' \
  -d '{
    "type": "A",
    "name": "app.example.com",
    "content": "10.0.0.50",
    "ttl": 300,
    "proxied": true,
    "use_private_routing": true
  }'

$ curl -X POST 'https://api.cloudflare.com/client/v4/zones/{zone_id}/spectrum/apps' \
  -H 'Authorization: Bearer <token>' \
  -H 'Content-Type: application/json' \
  -d '{
    "protocol": "tcp/22",
    "dns": { "type": "CNAME", "name": "ssh.example.com" },
    "origin_direct": ["tcp://10.0.0.50:22"],
    "virtual_network_id": "fab9ac85-491b-44c8-b7ae-dd44d4f4672e"
  }'

6 details in Everything

Cloudflare adds Cloudforce One threat intel in WAFWAFA reverse proxy that inspects HTTP/HTTPS traffic and blocks requests matching known attack patterns such as SQL injection or XSS, sitting between clients and a web application to filter malicious input before it reaches the server., Require Access Protection, IdPIdPA service that authenticates users and supplies identity information to other applications. It is the sign-in authority behind many single sign-on systems. Federation, and MCP ServerMCP ServerA process that exposes tools, resources, and prompts over Anthropic's Model Context Protocol, letting an AI agent call external APIs or local services through a standardised interface. Portal for agentic AI.

⤢

5 details in Everything

Cloudflare WAFWAFA reverse proxy that inspects HTTP/HTTPS traffic and blocks requests matching known attack patterns such as SQL injection or XSS, sitting between clients and a web application to filter malicious input before it reaches the server. gains live Threat Events fields, enabling rules that block by threat actor, targeted industry, and attack type.

⤢ ⤢ ⤢

$ any(cf.intel.ip.target_countries[*] == "FR") and any(cf.intel.ip.datasets[*] == "ddos")

$ any(cf.intel.ip.target_industries[*] == "Banking & Financial Services") and any(cf.intel.ip.attacker_names[*] == "BLACKBASTA")

8 details in Everything

Cloudflare AI Gateway adds dollar-based spend limits and identity-driven budgets with Cloudflare Access integration.

⤢ ⤢ ⤢

$ curl -X POST https://api.cloudflare.com/client/v4/accounts/<account_id>/ai-gateway/gateways/<gateway_id>/spend-limits \
  -H 'Authorization: Bearer <api_token>' \
  -H 'Content-Type: application/json' \
  -d '{
    "limit": 500,
    "currency": "USD",
    "window": "monthly",
    "window_type": "fixed",
    "attributes": {"team": "interns"},
    "on_exceed": "block"
  }'

7 details in Everything

Syft v1.45.0 adds ZapAddOns JAR cataloging, ingress-nginx binary detection, and MySQL Cluster/MySQL distinction.

$ git clone --branch v1.45.0 https://github.com/anchore/syft.git
# already have the repo? check out this version:
$ git checkout v1.45.0

$ syft registry.k8s.io/ingress-nginx/controller:v1.10.0 -o cyclonedx-json

3 details in Everything

Wiz Security Graph now ingests live runtime network signals to surface previously hidden attack pathsattack pathsA possible sequence of steps an attacker could take to reach a valuable target. It connects weaknesses, permissions and reachable systems rather than treating each finding separately. as critical Issues.

4 details in Everything

Wiz Runtime Sensor reaches GA for Windows, adding real-time threat detection, automated response, and runtime vulnerability validation across hybrid cloud.

7 details in Everything

Wiz launches Red Agent POV Series, showcasing its AI-powered pentester finding 5,500+ critical vulns across ~1,000 production environments.

4 details in Everything

Wiz launches Exposure Management Dashboard (Public Preview): a CTEMCTEMContinuous threat exposure management, a recurring process for finding, prioritizing and validating exploitable exposures. It connects discovery to fixes and checks whether those fixes work.-aligned single pane of glass for scoping, prioritizing, validating, and remediating attack pathsattack pathsA possible sequence of steps an attacker could take to reach a valuable target. It connects weaknesses, permissions and reachable systems rather than treating each finding separately. at AI speed.

8 details in Everything

Wiz adds native AI Code Scans, Red Agent adversarial validation, and Green Agent auto-remediation for AI-driven development.

7 details in Everything

Wiz adds Green Agent-powered remediation paths, unified ownership automation, CI/CD guardrails, and WizOS hardened images to accelerate patching.

8 details in Everything

Wiz Cloud Cost launches GA: unified cloud & AI spend visibility, optimization, and anomaly alerting across AWS, Azure, and GCP.

12 details in Everything

Wiz adds Red Agent (autonomous AI attacker) and Green Agent (AI remediation) alongside expanded Attack Surface Management for cloud, SaaSSaaSSoftware delivered over the internet and run on the provider's infrastructure, so the buyer pays a subscription instead of installing or operating anything locally., and AI environments.

5 details in Everything

Wiz API SPM is now GA, adding automated API discovery, exploitability testing, and toxic-combination prioritization across cloud environments.

7 details in Everything

MISP 2.5.41 adds event-index mass tagging, async server updates, new threat feeds, and a major Overmind UI rework.

$ git clone --branch v2.5.41 https://github.com/MISP/MISP.git
# already have the repo? check out this version:
$ git checkout v2.5.41

$ curl -s -H 'Authorization: <your-api-key>' -H 'Accept: application/json' -H 'Content-Type: application/json' -X POST https://<misp-instance>/servers/updateJSON -d '{"background": true}'

8 details in Everything

MISP 2.5.40 adds Overmind UI migration for Event views, ACLACLAn access control list, a set of rules allowing or denying operations for particular users, services or network addresses. It decides who can use a resource.-aware ATT&CKATT&CKA publicly maintained knowledge base by MITRE cataloguing adversary tactics, techniques, and procedures observed in real attacks, giving cyber tools a common taxonomy for classifying and correlating threat behavior. widget, and admin upgrade-error surfacing.

$ git clone --branch v2.5.40 https://github.com/MISP/MISP.git
# already have the repo? check out this version:
$ git checkout v2.5.40

4 details in Everything

MISP 2.5.39 ships a redesigned analyst dashboard, new threat-intel widgets, STIXSTIXStructured Threat Information Expression, a standard format for describing threats, indicators and their relationships. It helps organizations exchange threat intelligence in a usable form. TLP/patterning expansions, and a WebGL attack-flow globe.

$ git clone --branch v2.5.39 https://github.com/MISP/MISP.git
# already have the repo? check out this version:
$ git checkout v2.5.39

⤢ ⤢ ⤢ ⤢

10 details in Everything

CrowdStrike extends Continuous Identity to AI agents, AWS cloud infra, and unified non-human identitynon-human identityA non-human identity, an account or credential used by software or a device rather than a person. Service accounts, workload identities and automation tokens need their own permissions and lifecycle controls. visibility via Falcon Next-Gen Identity Security.

5 details in Everything

CrowdStrike joins OpenIDOpenIDAn open standard for decentralized authentication that lets a user prove their identity to a site via a trusted third-party provider, so tools can delegate login without managing credentials themselves. Foundation as Sustaining Member, enabling CAEP-based real-time identity signals across the Falcon ecosystem.

$ https://caep.dev

4 details in Everything

Elastic Security adds native Google Threat Intelligence integration with prebuilt dashboards, indicator match rules, and agentic VirusTotal enrichment.

⤢ ⤢

$ FROM logs-ti_google_threat_intelligence.* | WHERE threat.indicator.type == "ipv4-addr" AND google_threat_intelligence.threat_actor == "<actor-name>" | STATS count = COUNT() BY threat.indicator.ip, google_threat_intelligence.malware_family, google_threat_intelligence.score | SORT count DESC

7 details in Everything

InsightCloudSec 26.6.23 adds new insights, query filters, Kubernetes Scanner v5.1.0, and permission remediation improvements.

4 details in Everything

Rapid7 InsightVM 8.49.0 adds ASP.NET Core 10 fingerprinting and CISCISA nonprofit that publishes consensus-based configuration benchmarks and controls for hardening operating systems, cloud environments, and software; cyber tools use it as a ready-made compliance baseline. Microsoft Office Enterprise Benchmark v1.2.0 policy support.

2 details in Everything

Nexpose 8.49.0 adds ASP.NET Core 10 fingerprinting and CISCISA nonprofit that publishes consensus-based configuration benchmarks and controls for hardening operating systems, cloud environments, and software; cyber tools use it as a ready-made compliance baseline. Microsoft Office Enterprise Benchmark v1.2.0 policy support.

2 details in Everything

Rapid7 VM 8.48.0 adds Nutanix AHVAHVAcropolis Hypervisor, Nutanix's built-in bare-metal hypervisor for running VMs on their hyper-converged infrastructure, giving cyber tools a tightly integrated compute and storage layer without a separate hypervisor license. fingerprinting, TLS 1.3 named-group reporting, and richer credential logging.

3 details in Everything

Nexpose 8.48.0 adds TLS 1.3 named-group reporting, Nutanix AHVAHVAcropolis Hypervisor, Nutanix's built-in bare-metal hypervisor for running VMs on their hyper-converged infrastructure, giving cyber tools a tightly integrated compute and storage layer without a separate hypervisor license. fingerprinting, and richer credential success logging.

3 details in Everything

InsightCloudSec 26.6.9 adds ECSECSElastic Common Schema, an open specification maintained by Elastic that defines a standard field naming convention for log and event data, letting cyber tools share and correlate records without per-source field mapping. insights, enhanced logging, and EventBridgeEventBridgeAWS's managed event bus service that routes events between cloud services, SaaS apps, and custom code using rules and targets, letting tools react to state changes without polling or managing message infrastructure. filtering for cloud security.

3 details in Everything

InsightCloudSec 26.6.2 adds new insights, query filters, bot action improvements, and enhanced compliance pack filtering.

4 details in Everything

DefectDojo 3.0 ships a full Tailwind UI rebuild, new parsers, localization, and removes EOL features.

$ git clone --branch 3.0.0 https://github.com/DefectDojo/django-DefectDojo.git
# already have the repo? check out this version:
$ git checkout 3.0.0

18 details in Everything

Claude Code v2.1.193 adds shell classifier controls, OTelOTelAn 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. response logging, live path autocomplete, and memory-pressure reaping.

$ git clone --branch v2.1.193 https://github.com/anthropics/claude-code.git
# already have the repo? check out this version:
$ git checkout v2.1.193

$ claude config set autoMode.classifyAllShell true

$ OTEL_LOG_ASSISTANT_RESPONSES=1 claude

$ CLAUDE_CODE_DISABLE_BG_SHELL_PRESSURE_REAP=1 claude

9 details in Everything

Claude Code v2.1.191 adds /rewind for post-clear conversation recovery and cuts streaming CPUCPUThe main general-purpose processor in a computer. It runs application instructions and coordinates work with memory and other hardware. usage by ~37%.

$ git clone --branch v2.1.191 https://github.com/anthropics/claude-code.git
# already have the repo? check out this version:
$ git checkout v2.1.191

$ /rewind

7 details in Everything

Claude Code v2.1.187 adds credential sandboxingsandboxingRunning 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., org model restrictions, mouse support in fullscreen, and UX improvements to /btw, /plugin, and /install-github-app.

$ git clone --branch v2.1.187 https://github.com/anthropics/claude-code.git
# already have the repo? check out this version:
$ git checkout v2.1.187

$ claude config set sandbox.credentials false

6 details in Everything

Claude Code v2.1.186 adds CLI 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. auth commands, bash auto-response, AWS credential refresh, and subagent permission surfacing.

$ git clone --branch v2.1.186 https://github.com/anthropics/claude-code.git
# already have the repo? check out this version:
$ git checkout v2.1.186

$ claude mcp login my-server --no-browser

$ CLAUDE_CODE_RETRY_WATCHDOG=1 claude --print 'run the full test suite and fix any failures'

11 details in Everything

Claude Code v2.1.183 adds auto-mode guardrails against destructive commands, attribution controls, and improved /config UX.

$ git clone --branch v2.1.183 https://github.com/anthropics/claude-code.git
# already have the repo? check out this version:
$ git checkout v2.1.183

$ /config --help

6 details in Everything

Claude Code v2.1.181 adds inline /config setting, Apple Events sandbox opt-in, and presence-based push notification suppression.

$ git clone --branch v2.1.181 https://github.com/anthropics/claude-code.git
# already have the repo? check out this version:
$ git checkout v2.1.181

$ /config thinking=false

$ CLAUDE_CLIENT_PRESENCE_FILE=~/.at-desk claude

8 details in Everything

Claude Code v2.1.178 adds parameter-scoped permission rules, nested skills/config resolution, and streamlined agent teams.

$ git clone --branch v2.1.178 https://github.com/anthropics/claude-code.git
# already have the repo? check out this version:
$ git checkout v2.1.178

$ claude --disallow-tool 'Agent(model:opus)' "Refactor the auth module"

$ CLAUDE_CODE_EXPERIMENTAL_AGENT_TEAMS=1 claude "Use the Agent tool with name:reviewer to audit my PR diff"

8 details in Everything

Codex CLI now enforces enterprise marketplace source policy at runtime, blocking inactive or disallowed plugins.

$ git clone --branch rust-v0.143.0-alpha.29 https://github.com/openai/codex.git
# already have the repo? check out this version:
$ git checkout rust-v0.143.0-alpha.29

3 details in Everything

Adds experimental environment/info app-server RPCRPCRemote Procedure Call, a protocol that lets a program execute a function on another machine as if it were local, giving cyber tools a structured way to communicate across processes or hosts without managing raw sockets. to discover shell and working directory for named execution environments.

$ git clone --branch rust-v0.143.0-alpha.28 https://github.com/openai/codex.git
# already have the repo? check out this version:
$ git checkout rust-v0.143.0-alpha.28

{
  "id": 42,
  "jsonrpc": "2.0",
  "method": "environment/info",
  "params": { "name": "<environment-name>" }
}

2 details in Everything

Codex CLI adds service tier and reasoning effort to OTELOTELAn 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. telemetry and surfaces rollout budget exhaustion to users.

$ git clone --branch rust-v0.143.0-alpha.27 https://github.com/openai/codex.git
# already have the repo? check out this version:
$ git checkout rust-v0.143.0-alpha.27

12 details in Everything

Codex CLI now surfaces service tier and reasoning effort in OTELOTELAn 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. telemetry and exposes sandbox state in debug sandbox commands.

$ git clone --branch rust-v0.143.0-alpha.26 https://github.com/openai/codex.git
# already have the repo? check out this version:
$ git checkout rust-v0.143.0-alpha.26

7 details in Everything

Codex CLI adds singular --permission-profile flag, provider-aware model fallback, current-time reminder config, and exec-server lifecycle metrics.

$ git clone --branch rust-v0.143.0-alpha.25 https://github.com/openai/codex.git
# already have the repo? check out this version:
$ git checkout rust-v0.143.0-alpha.25

$ codex sandbox --permission-profile strict

7 details in Everything

Adds HTTP 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. server support with OAuth, server-attributed error telemetry, and raised token-budget message limits.

$ git clone --branch rust-v0.143.0-alpha.22 https://github.com/openai/codex.git
# already have the repo? check out this version:
$ git checkout rust-v0.143.0-alpha.22

5 details in Everything

Codex CLI gains Ultra reasoning effort, ChatGPT session auth for 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. servers, remote plugin catalog polish, and world-state persistence.

$ git clone --branch rust-v0.143.0-alpha.21 https://github.com/openai/codex.git
# already have the repo? check out this version:
$ git checkout rust-v0.143.0-alpha.21

$ codex mcp add --transport http https://chatgpt.com/your-mcp-endpoint
# Then set use_chatgpt_auth: true in ~/.codex/config.json under that server's entry

6 details in Everything

ClinePass provider is now enabled for all users, no longer gated behind a feature flag.

$ git clone --branch v4.0.3 https://github.com/cline/cline.git
# already have the repo? check out this version:
$ git checkout v4.0.3

1 detail in Everything

Cline v4.0.2 adds DeepSeek reasoning effort control (including xhigh) and a revamped ClinePass provider UX.

$ git clone --branch v4.0.2 https://github.com/cline/cline.git
# already have the repo? check out this version:
$ git checkout v4.0.2

2 details in Everything

Cline CLI v3.0.31 adds subscription cost visibility, ClinePass prompts, marketplace uninstall, and one-shot quoted-prompt enforcement.

$ git clone --branch cli-v3.0.31 https://github.com/cline/cline.git
# already have the repo? check out this version:
$ git checkout cli-v3.0.31

6 details in Everything

Cline v4.0.0 ships SDK-backed runtime, ClinePass, a Customize marketplace, Plugins, queued prompts, and new model providers.

$ git clone --branch v4.0.0 https://github.com/cline/cline.git
# already have the repo? check out this version:
$ git checkout v4.0.0

14 details in Everything

Cline CLI v3.0.30 adds SAP AI Core provider, token-cost status bar, org error messages, and proxy-aware networking.

$ git clone --branch cli-v3.0.30 https://github.com/cline/cline.git
# already have the repo? check out this version:
$ git checkout cli-v3.0.30

7 details in Everything

OpenHands 1.39.0 adds multi-model BYOKBYOKBring Your Own Key — a cloud deployment model where the customer supplies and controls the encryption keys rather than the provider, so sensitive data stays under the customer's cryptographic authority. discovery, 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. server config for ACP agents, Slack attachment context, and org-management improvements.

$ git clone --branch cloud-1.39.0 https://github.com/OpenHands/OpenHands.git
# already have the repo? check out this version:
$ git checkout cloud-1.39.0

23 details in Everything

OpenHands 1.8.0 adds LLMLLMA 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. profiles, sub-agent delegation, sandbox grouping strategy, and an ACP agent UI.

$ git clone --branch 1.8.0 https://github.com/OpenHands/OpenHands.git
# already have the repo? check out this version:
$ git checkout 1.8.0

4 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.

$ git clone --branch v1.38.0 https://github.com/aaif-goose/goose.git
# already have the repo? check out this version:
$ git checkout v1.38.0

$ goose review --turn-limit 10

$ GOOSE_SERVER__SECRET_KEY=mysecretkey goose serve

$ GOOSE_MAX_TOOL_RESPONSE_SIZE=8192 GOOSE_DISABLE_TOOL_CALL_SUMMARY=true goose run --recipe my-recipe.yaml

32 details in Everything

Goose v1.37.0 adds 9+ new providers, PreToolUse hooks, goose review, /model switching, skills listing, and encrypted session sharing.

$ git clone --branch v1.37.0 https://github.com/aaif-goose/goose.git
# already have the repo? check out this version:
$ git checkout v1.37.0

$ goose review

$ /model claude-opus-4-5

$ goose skills list

$ GOOSE_MAX_TOOL_RESPONSE_SIZE=50000 goose run --recipe my-recipe.yaml

37 details in Everything

OpenCode v1.17.10 adds 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. resource tools, a new api CLI command, --mini mode, Gemini audio/video, and a V2 plugin API.

$ git clone --branch v1.17.10 https://github.com/anomalyco/opencode.git
# already have the repo? check out this version:
$ git checkout v1.17.10

$ opencode --mini

16 details in Everything

OpenCode v1.17.9 adds GLM-5.2 high/max thinking variants and improves prompt-cache efficiency for follow-up messages.

$ git clone --branch v1.17.9 https://github.com/anomalyco/opencode.git
# already have the repo? check out this version:
$ git checkout v1.17.9

1 detail in Everything

OpenCode v1.17.8 adds a Home tab toggle and faster session timelines in the desktop app.

$ git clone --branch v1.17.8 https://github.com/anomalyco/opencode.git
# already have the repo? check out this version:
$ git checkout v1.17.8

3 details in Everything

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. servers now receive the current workspace as a client root; SDK clients refresh model availability when integrations change.

$ git clone --branch v1.17.7 https://github.com/anomalyco/opencode.git
# already have the repo? check out this version:
$ git checkout v1.17.7

2 details in Everything

Adds external browser OAuth for Snowflake Cortex and brings v2 visibility settings to the desktop web app.

$ git clone --branch v1.17.5 https://github.com/anomalyco/opencode.git
# already have the repo? check out this version:
$ git checkout v1.17.5

3 details in Everything

OpenCode v1.17.4 adds 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. server log surfacing, cwd support for local MCP servers, connector auth, and new v2 session API endpoints.

$ git clone --branch v1.17.4 https://github.com/anomalyco/opencode.git
# already have the repo? check out this version:
$ git checkout v1.17.4

$ opencode --mcp-server '{"type":"local","command":"node","args":["server.js"],"cwd":"tools/mcp"}'

4 details in Everything

OpenCode v1.17.1 adds usage descriptions and hidden-from-autocomplete support for references, plus a dedicated /new-session desktop route.

$ git clone --branch v1.17.1 https://github.com/anomalyco/opencode.git
# already have the repo? check out this version:
$ git checkout v1.17.1

3 details in Everything

Crush v0.80.0 adds auto-detection of local model context windowscontext windowsThe 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 a llamacpp provider enricher.

$ git clone --branch v0.80.0 https://github.com/charmbracelet/crush.git
# already have the repo? check out this version:
$ git checkout v0.80.0

3 details in Everything

Bang Mode gains live output streaming, cancellation, copy-to-clipboard, and string-prefix invocation in v0.79.0.

$ git clone --branch v0.79.0 https://github.com/charmbracelet/crush.git
# already have the repo? check out this version:
$ git checkout v0.79.0

4 details in Everything

Crush v0.78.0 adds bang mode for inline shell execution, local model auto-discovery, Fireworks provider support, and improved clipboard handling.

$ git clone --branch v0.78.0 https://github.com/charmbracelet/crush.git
# already have the repo? check out this version:
$ git checkout v0.78.0

⤢ ⤢

$ !ls -la /etc/cron.d

{
  "providers": {
    "ollama": {
      "name": "Ollama",
      "base_url": "http://localhost:11434/v1/",
      "type": "ollama"
    }
  }
}

4 details in Everything

Crush v0.77.0 adds global user-level context files and named hooks for easier hook management.

$ git clone --branch v0.77.0 https://github.com/charmbracelet/crush.git
# already have the repo? check out this version:
$ git checkout v0.77.0

$ mkdir -p ~/.config/crush
cat >> ~/.config/crush/CRUSH.md <<'EOF'
Always refer to me as Shirley. Default to Jujutsu (jj) for version-control commands unless the project already uses plain git.
EOF

{
  "hooks": {
    "PreToolUse": [
      {
        "name": "lint-before-bash",
        "matcher": "bash",
        "command": "./hooks/run-lint.sh"
      }
    ]
  }
}

2 details in Everything

Crush v0.76.0 adds monorepo skill discovery and shows all providers in crush models

$ git clone --branch v0.76.0 https://github.com/charmbracelet/crush.git
# already have the repo? check out this version:
$ git checkout v0.76.0

$ crush models

2 details in Everything

Cursor's new Customize page unifies plugins, skills, MCPs, rules, and hooks with a team marketplace leaderboard and prebuilt canvases.

5 details in Everything

Cursor Automations gains /automate skill, Slack emoji trigger, five new GitHub triggers, and computer use support.

$ /automate When a PR review comment is left, read the comment and push a commit that addresses the feedback.

8 details in Everything

Cursor adds cloud dev environment snapshots and /in-cloud subagents for parallel, isolated agent work in the Agents Window.

⤢ ⤢

$ /in-cloud Fix the failing integration tests in the CI pipeline and open a PR when done

6 details in Everything

Bugbot is 3x faster and 10% more accurate; adds pre-push /review command and incremental PR review mode.

⤢ ⤢

$ /review

$ /review-security

6 details in Everything

Kiro IDE 1.0 adds Agent Focus mode, capability-based permissions, custom Markdown-defined agents, NL hook creation, and dockable chat tabs.

# .kiro/agents/code-reviewer.md
---
tools: read, web
mcp:
  - server: github
permissions:
  deny: shell
---
You are a strict code reviewer. Focus on security, clarity, and test coverage.

6 details in Everything

Kiro adds compact tool card previews for sub-agent calls.

Kiro adds Automations: schedule recurring AI-driven tasks that run in sandboxes and open pull requests automatically.

⤢

4 details in Everything

Kiro CLI v3 Early Access brings spec-driven development, capability-based permissions, enhanced hooks, and tag-based agent config to the terminal.

$ kiro-cli --v3

5 details in Everything

Kiro adds /goal loops, mid-turn queue steering, and enriched /rewind previews for more agentic control.

$ /goal --max 8 "Implement OAuth2 login flow with tests passing and no lint errors"

4 details in Everything

Kiro Web gains full GitLab support and brings the Specs planning workflow to the browser.

6 details in Everything

Kiro launches Pro Max tier at $100/month with 5,000 credits and full feature access.

1 detail in Everything

Kiro adds transcript export, terminal window titles, launch-time effort control, and persistent model/effort preferences.

$ /transcript save --json

$ kiro-cli chat --effort low

4 details in Everything

Kiro Web lets you start sessions without a GitHub repo and switch to Autonomous mode mid-session.

3 details in Everything

Gemini CLI v0.47.0 adds Antigravity CLI migration commands and promotes browser agent to stable.

$ git clone --branch v0.47.0 https://github.com/google-gemini/gemini-cli.git
# already have the repo? check out this version:
$ git checkout v0.47.0

3 details in Everything

CrewAI 1.15.0 adds declarative Flow definitions with crew/agent actions, each composite steps, conditional expressions, and DMN mode support.

$ git clone --branch 1.15.0 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout 1.15.0

12 details in Everything

CrewAI 1.14.7 adds a chat API, Snowflake Cortex LLMLLMA 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., pluggable memory/RAGRAGRetrieval-augmented generation: looking up relevant documents and adding them to the prompt before the model answers. It grounds responses in private or current data without retraining. backends, richer LLM event metadata, and route-aware Flow decorators.

$ git clone --branch 1.14.7 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout 1.14.7

from crewai.flow.flow import Flow, listen, router, start

class TriageFlow(Flow):
    @start()
    def ingest(self):
        self.state.severity = classify_alert()

    @router(ingest)
    def route_by_severity(self):
        if self.state.severity == "critical":
            return "escalate"
        return "log_only"

    @listen("escalate")
    def escalate(self):
        ...

    @listen("log_only")
    def log_only(self):
        ...

from crewai import Agent, LLM

analyst = Agent(
    role="Data Analyst",
    goal="Summarize query results from Snowflake",
    backstory="Expert in SQL and data warehousing.",
    llm=LLM(model="snowflake/snowflake-arctic-instruct"),
)

9 details in Everything

CrewAI 1.15.1a1 adds TUITUIA 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. button telemetry, enforces explicit project definitions, and auto-opens the deployment page after CLI deploy.

$ git clone --branch 1.15.1a1 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout 1.15.1a1

3 details in Everything

CrewAI 1.14.8a5 enables declarative refs to work seamlessly across both Flows and Crews.

$ git clone --branch 1.14.8a5 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout 1.14.8a5

1 detail in Everything

CrewAI 1.14.8a4 adds conversational flow support in the CLI TUITUIA 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..

$ git clone --branch 1.14.8a4 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout 1.14.8a4

1 detail in Everything

CrewAI 1.14.8a3 adds declarative Flow loading, unified CLI commands, router-as-start support, and typed tool output schemas.

$ git clone --branch 1.14.8a3 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout 1.14.8a3

from crewai.flow.flow import Flow, router, listen, start

class TriageFlow(Flow):
    @router()
    @start()
    def triage(self):
        if self.state.priority == 'high':
            return 'urgent'
        return 'normal'

    @listen('urgent')
    def handle_urgent(self): ...

    @listen('normal')
    def handle_normal(self): ...

6 details in Everything

CrewAI 1.14.8a2 adds single-agent actions inside Flow definitions and 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. expression validation at load time.

$ git clone --branch 1.14.8a2 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout 1.14.8a2

2 details in Everything

CrewAI 1.14.8a1 adds optional if expressions to each.do loop steps in Flows.

$ git clone --branch 1.14.8a1 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout 1.14.8a1

1 detail in Everything

CrewAI 1.14.8a introduces JSON-first crew definitions, FlowDefinition actions, DMN mode, and a new crewai run --definition CLI flag.

$ git clone --branch 1.14.8a https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:
$ git checkout 1.14.8a

$ crewai run --definition

9 details in Everything

LangGraph CLI now supports running the dev server over HTTPSHTTPSHTTP carried over an encrypted TLS connection. It protects traffic in transit and lets a browser verify the server's certificate. with custom certificate and key files.

$ git clone --branch cli==0.4.29 https://github.com/langchain-ai/langgraph.git
# already have the repo? check out this version:
$ git checkout cli==0.4.29

$ langgraph dev --certfile cert.pem --keyfile key.pem

1 detail in Everything

LangGraph 1.2.3 adds v3 streaming, 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. transport, and tool-dispatched subagent naming to RemoteGraph.