#66th of 19 in Security
failproofai
Policy enforcement and run tracing for coding agent CLIs and gateways
- Stars
- 8.5k
- License
- custom license
- Last commit
- Nov 2026
- Last release
- Nov 2026
- Language
- TypeScript
204 on 21 Apr 20268,531 on 10 Nov 2026
From GitHub's daily star counts.
Checked on GitHub.
Overview
Failproof AI installs hooks into 12 agent harnesses (Claude Code, Codex, Copilot CLI, Cursor Agent, OpenCode, Hermes, OpenClaw and others). It blocks tool calls such as rm -rf, sudo, force pushes and .env reads before they run, using 40 built-in policies plus custom JS policies. A local dashboard on localhost:8020 shows session history, hook decisions and an offline audit of past runs.
Who it is for: Teams running coding agents who want guardrails and audit trails
Strengths
- Pre-execution blocking of tool calls verified on all 12 supported harnesses
- Custom policies are JS files in .failproofai/policies/, committed and shared with the team
- Local dashboard and audit read existing history with no account or upload
- Policy API is small: allow, deny, instruct
Weaknesses
- License is MIT with Commons Clause, not OSI open source; resale needs a separate agreement
- Setup selects no policies; only block-failproofai-commands is active until a pack is added
- Turn-end gates work on 8 of 12 harnesses; capabilities differ per harness
- Fleet-wide observability and cluster self-hosting are hosted or Enterprise offerings
What it needs
- no GPU
- Docker
- Needs Node.js (npm)
- port 8020
Quick answers
Is failproofai open source?
Its code is public on GitHub under its own license terms; read them before you use it at work.
Can you run failproofai with Docker?
Yes. Its repository has a Docker setup: a Dockerfile, a Compose file or a published image.
Does failproofai need a GPU?
No, it runs without one.
Also in Security
See all 19| Rank | Project | Score |
|---|---|---|
| 1 | shannonAutonomous pentester that exploits web apps and APIs using source analysis | 91 out of 100 |
| 2 | ifixaiPython CLI that audits AI agents for policy, access and reliability failures | 87 out of 100 |
| 3 | StrixAutonomous AI pentesting agents that validate vulnerabilities with working exploits | 84 out of 100 |
| #4 | pentagiAutonomous multi-agent penetration testing system with sandboxed security tools | 77 out of 100 |
| #5 | ai-infra-guardRed-teaming platform for AI infra CVEs, MCP servers, agent skills and jailbreaks | 75 out of 100 |