← Browse
rrsi-policy
grapefruit0205 ★ 0
rrsi-policy is a runtime policy engine that screens Claude Code harness edits (CLAUDE.md, skills, hooks, MCP) through a separate critic process before they land, using deterministic routing and measurement-based verdicts adapted from google-research/rrsi. It enforces policies via PreToolUse and PostToolUse hooks, maintains a decision ledger, and combines secret-regex checks for fast rejections with LLM-based critique for configuration changes.
Install
> /plugin marketplace add grapefruit0205/rrsi-policy
> /plugin install rrsi-policy
What it's made of
2 hooks
- Commands
- 0
- Agents
- 0
- Skills
- 0
- MCP servers
- 0
- Hooks
- 2 · PostToolUse, PreToolUse
What it needs & plugs into
- API keys
- none
- Paid services
- Anthropic
- External tools
python3- Talks to
- nothing external detected
Facts extracted from the plugin's files. Prose generated by claude-haiku-4-5-20251001.
Is this your plugin?
Claim it to keep the card accurate and enter the weekly contest. Requires signing in as the GitHub owner (grapefruit0205).