arai
CLAUDE.md7 starsChanged 6 months agoApache-2.0View on GitHub
Run arai init to register Arai's command hooks in .claude/settings.json. Trust the project in Claude Code, then try a harmless action covered by a Block rule. The tool call should be blocked and arai audit should show the decision. On Windows the registered hook starts PowerShell with an encoded command. Git Bash and PowerShell can both invoke it, and the executable path remains literal even when it contains spaces or shell metacharacters. Re-run arai init to refresh an existing registration.
In this repository9 other files it gives its agents.
- AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- compositionSkill
- contract-derivationSkill
- design-docSkill
- dispatcherSkill
- leaf-implementationSkill
- orchestratorSkill
- verifierSkill
The file
We split this claude.md into the rules it gives an agent
1 rule. About 0 are common, 0 are distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈0
Distinctive · estimate
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 3
Run arai init to register Arai's command hooks in .claude/settings.json.
≈ “Common” is an estimate: at least 1 in 5 of the 27,641 CLAUDE.md files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Claude Code integration
Run arai init to register Arai's command hooks in .claude/settings.json.
Trust the project in Claude Code, then try a harmless action covered by a Block
rule. The tool call should be blocked and arai audit should show the decision.
On Windows the registered hook starts PowerShell with an encoded command.
Git Bash and PowerShell can both invoke it, and the executable path remains
literal even when it contains spaces or shell metacharacters. Re-run arai
init to refresh an existing registration.
Claude Code Mods
Mods run inside Claude Code. An installed Mod that handles tool.check can
approve a call after a PreToolUse hook denies it when the session has neither
managed settings nor a Team or Enterprise /login credential. Under either
managed settings or that login, Claude Code preserves the hook's deny even
when the hook comes from project settings. An API key or long-lived
CLAUDE_CODE_OAUTH_TOKEN does not count as a Team or Enterprise /login.
Claude Code permission precedence
Claude Code enables Mods by default starting with 2.1.286. Administrators can
set allowManagedModsOnly in managed settings to limit which Mods load.
Claude Code managed Mods
Hook failures
Claude Code 2.1.295 and later support onFailure: "block" for command hooks.
That option controls failures such as a missing binary or a timeout. Arai's
registered Claude hook uses the host's default failure policy and returns a
JSON deny when it evaluates a Block rule.
Claude Code hook failure behavior
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

