# handrail-ai

> Evidence-based guardrails for AI coding agents: scan risky agent setups (MCP auto-approve, bypassPermissions, leaked keys), stop dangerous tool calls live, catch unverified 'tests pass' claims.

Latest version **0.3.0** (published 2026-11-19) · MIT license · 0 weekly downloads

_The full health report (score, vulnerabilities, README, dependency tree, version history) is being computed; this document will grow once it lands._

## Install

```sh
npm install handrail-ai
pnpm add handrail-ai
yarn add handrail-ai
bun add handrail-ai
```

## Facts

| | |
|---|---|
| Version | 0.3.0 |
| Published | 2026-11-19 |
| First published | 2026-11-19 |
| Weekly downloads | 0 |
| License | MIT |
| TypeScript types | none |
| Module format | CommonJS |
| Node | >=18 |
| Dependencies | 0 |
| Unpacked size | 101.6 KB |
| Author | Yeswanth Varma Kanumuri |
| Maintainers | kanumuri |
| Keywords | ai, agents, claude-code, mcp, security, guardrails, ci, sarif, prompt-injection, cursor, copilot |

## Links

- npm: https://www.npmjs.com/package/handrail-ai
- Repository: https://github.com/kanumuri9593/handrail
- Homepage: https://handrail-axa.pages.dev
- npm.io page: https://npm.io/package/handrail-ai

---
_Source: https://npm.io/package/handrail-ai · Machine-readable twin of the npm.io package page. Health data is recomputed on every publish._
