Install Claude Code before setup. The integration key is claude. This integration is declared multi-install safe. See the shared multi-install rules. For an existing Spec Kit project, use specify integration install claude. See integration management for activation, upgrades, and removal. Skills are installed as .claude/skills/speckit-<command>/SKILL.md. Invoke them in Claude's chat, for example /speckit-specify or /speckit-bug-assess for the installed bug extension. By default every Spec Kit skill installed for Claude Code is model-invocable, so its description is always in Claude's context…
Steady+42 stars this week
No other project uses it yet
149k12d agoReviewMountCreate a free account to mount
awf (Agentic Workflow Firewall, package @github/awf) is a CLI that wraps any command in a sandboxed Docker network. It provides L7 (HTTP/HTTPS) egress control using Squid proxy, restricting network access to a whitelist of approved domains while giving the agent access to the host workspace and selected system paths via chroot and selective bind mounts. The system is orchestrated by src/cli.ts and managed by src/docker-manager.ts. Squid, the primary agent, and the general API proxy are the baseline services; private-repository enclaves…
15748d agoReviewMountCreate a free account to mount
gh-aw-threat-detection is the threat detection component for GitHub Agentic Workflows (gh-aw). It is a Go CLI (threat-detect) that analyzes artifacts produced by AI agents — prompts, agent output, git patches, comment memory — and decides whether to allow or block the downstream safe-outputs job. The component runs in two main contexts: It detects three categories: prompt injection, secret leak, and malicious patch, and emits a strict JSON contract. Run these from the repo root. They are the canonical entry points…
1557d agoReviewMountCreate a free account to mount
gh-aw-threat-detection is the threat detection component for GitHub Agentic Workflows (gh-aw). It is a Go CLI (threat-detect) that analyzes artifacts produced by AI agents — prompts, agent output, git patches, comment memory — and decides whether to allow or block the downstream safe-outputs job. The component runs in two main contexts: It detects three categories: prompt injection, secret leak, and malicious patch, and emits a strict JSON contract. Run these from the repo root. They are the canonical entry points…
A file Claude Code reads at the start of every session. It holds the commands, conventions and warnings the agent needs for this project.
Where does it go?
At the repository root. Claude Code also reads CLAUDE.md files in subdirectories when it works there.
What should it contain?
Build and test commands, the project's layout, conventions that aren't obvious from the code, and mistakes to avoid. Short files tend to work better than long ones.
CLAUDE.md or AGENTS.md?
Claude Code reads CLAUDE.md; most other agents read AGENTS.md. Many projects keep one and point the other at it.
GitHubCLAUDE.md