Skip to content
Back to search

pegasi

MarketplaceWorks with Claude CodeGitHub

Stop AI agents from doing things you didn't ask for.

Safety scan

A static text scan, not a guarantee. Select the badge to see what was scanned and every finding.

Checked today
Badge for your README
findsafeskills safety rating

Markdown

[![findsafeskills safety rating](/api/listings/pegasi-ai-reins/badge.svg)](/listings/pegasi-ai-reins)

HTML

<a href="/listings/pegasi-ai-reins"><img src="/api/listings/pegasi-ai-reins/badge.svg" alt="findsafeskills safety rating" /></a>

Install

Install instructions from the README

As written by the repo's author. findsafeskills has not verified these steps.

npm install -g @pegasi-ai/reins
reins init

What was scanned

The files the safety scan read, as written by the repository's author and shown as plain text. Lines with a finding are marked.

.claude-plugin/marketplace.json · Manifest · 24 lines

View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)

  1. {
  2. "name": "pegasi",
  3. "owner": {
  4. "name": "Kevin Wu",
  5. "email": "kevin@usepegasi.com"
  6. },
  7. "metadata": {
  8. "description": "Security controls for AI agents.",
  9. "version": "0.0.13"
  10. },
  11. "plugins": [
  12. {
  13. "name": "reins",
  14. "source": "./plugin",
  15. "description": "Deterministic policy enforcement, OWASP ASI10 scanning, and audit trails for Claude Code and MCP-compatible agents.",
  16. "version": "0.0.13",
  17. "author": { "name": "Kevin Wu", "email": "kevin@usepegasi.com" },
  18. "homepage": "https://reins.sh/docs",
  19. "repository": "https://github.com/pegasi-ai/reins",
  20. "license": "Apache-2.0",
  21. "keywords": ["security", "governance", "ai-agents", "hooks", "audit", "mcp"]
  22. }
  23. ]
  24. }
README.md · README · 81 lines

View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)

  1. <div align="center">
  2. <img src="./public/reins_logo.png" alt="Reins Logo" width="360"/>
  3. <h1>🪢 Reins</h1>
  4. <p><strong>Security controls for AI agents.</strong></p>
  5. <p>
  6. <a href="https://github.com/pegasi-ai/reins">github.com/pegasi-ai/reins</a>
  7. </p>
  8. <p>
  9. <a href="https://www.npmjs.com/package/clawreins"><img src="https://img.shields.io/npm/v/clawreins.svg" alt="npm"></a>
  10. <a href="https://www.npmjs.com/package/clawreins"><img src="https://img.shields.io/npm/dt/clawreins.svg" alt="downloads"></a>
  11. <a href="https://www.apache.org/licenses/LICENSE-2.0"><img src="https://img.shields.io/badge/License-Apache%202.0-blue.svg" alt="License: Apache 2.0"></a>
  12. <a href="http://www.typescriptlang.org/"><img src="https://img.shields.io/badge/%3C%2F%3E-TypeScript-%23007ACC.svg" alt="TypeScript"></a>
  13. <img src="https://img.shields.io/badge/node-%3E%3D18.0.0-brightgreen" alt="Node.js >= 18.0.0">
  14. </p>
  15. </div>
  16. > In Greek myth, Athena gave Bellerophon the golden bridle — reins included — that let him guide Pegasus. Reins applies the same idea to AI agents: raw power is not enough — what matters is making it controllable.
  17. Reins enforces deterministic security policies on every agent action, scans your configs for OWASP ASI10 vulnerabilities, and tracks drift over time. Policies evaluate in under 50ms. Works with Claude Code PreToolUse and PostToolUse hooks, OpenClaw, and any MCP-compatible agent.
  18. ## Quickstart
  19. ```bash
  20. npm install -g @pegasi-ai/reins
  21. reins init
  22. ```
  23. ## Claude Code Skill
  24. Install the Reins skill to give Claude Code awareness of your security posture:
  25. ```bash
  26. mkdir -p ~/.claude/skills/reins
  27. curl -o ~/.claude/skills/reins/SKILL.md \
  28. https://raw.githubusercontent.com/pegasi-ai/reins/main/.claude/skills/reins/SKILL.md
  29. ```
  30. Or clone the repo — the skill is included at `.claude/skills/reins/` automatically.
  31. ## Demo
  32. ![Reins demo](./public/reins-demo.gif)
  33. An OpenClaw agent tries to bulk-delete 4,382 Gmail messages. Reins blocks it before execution.
  34. ## What Reins does
  35. - **Prevent** — Block destructive actions before execution. Score irreversibility. Detect risky browser state.
  36. - **Pause** — Route high-impact actions through terminal or messaging approval flows. Require explicit `CONFIRM-*` tokens for catastrophic operations.
  37. - **Prove** — Preserve an immutable audit trail of every decision, approval, and block.
  38. ## Security guarantees
  39. - **Zero Trust** — every action evaluated before execution
  40. - **Synchronous** — agent cannot proceed until the hook exits
  41. - **No network in the hot path** — policies cached locally, enforced offline
  42. - **Fail-closed** — any unhandled hook error blocks the action
  43. - **Immutable audit** — append-only JSONL at `~/.openclaw/reins/decisions.jsonl`
  44. ## Documentation
  45. Full docs at [reins.sh/docs](https://reins.sh/docs):
  46. - [Getting Started](https://reins.sh/docs/getting-started)
  47. - [How It Works](https://reins.sh/docs/how-it-works)
  48. - [Security Policies](https://reins.sh/docs/policies)
  49. - [CLI Reference](https://reins.sh/docs/cli)
  50. - [Security Scan](https://reins.sh/docs/scan)
  51. - [Reins Cloud](https://reins.sh/docs/reins-cloud)
  52. - [Use as a Library](https://reins.sh/docs/library)
  53. - [Architecture](https://reins.sh/docs/architecture)
  54. ## Contributing
  55. PRs welcome. See [CONTRIBUTING.md](CONTRIBUTING.md).
  56. ## License
  57. Apache 2.0 — see [LICENSE](LICENSE).

Ratings and comments

Sign in to rate this skill.

Sign in

Free to use. If it’s helped, pass it on.