skills.rest
FailproofAIFailproofAIOfficial·3 Agent Skills Included

failproofai

Monitor AI agents and block dangerous actions before they run

Records every AI agent run and blocks dangerous tool calls like sudo, force pushes, and secret file reads before they execute. Works across Claude Code, Codex, Copilot, Cursor, and other coding agents, plus custom agents via a Python SDK. Eliminates blind spots in agent behavior with session history, audits, and enforceable team-wide safety policies.
1,722|417
↗
npx skills add FailproofAI/failproofai --all -g -y
Available:

Gives AI coding agents strict operating rules for this repository, including testing requirements, hook configuration across Claude Code, Codex, Copilot, and Cursor, and validation steps before any change is pushed.

All Skills in This Repository (3)

Pure Emerald Level Indicators
📦 In Repo
FailproofAIFailproofAI

mintlify

Automate MDX documentation site maintenance and configuration validation with the Mintlify CLI.

Official
Intermediate
📦 In Repo
FailproofAIFailproofAI

failproofai-sdk

Instrument Python AI agents to emit structured telemetry events for the Failproof AI platform.

Official
Advanced
📦 In Repo
FailproofAIFailproofAI

fp-cloud-cli

Inspect agent telemetry and operate a FailproofAI Cloud deployment via the fp CLI.

Official
Advanced

Frequently Asked Questions

FAQPage Schema
How to install FailproofAI?▼

Run `npx skills add FailproofAI/failproofai --all -g -y` in your terminal to install all skills in this suite globally.

What does FailproofAI block?▼

It blocks dangerous agent actions like sudo commands, recursive file deletion, force pushes, reading .env files, and unreviewed infrastructure changes, using 39 built-in policies plus your own custom rules.

Which AI coding agents does FailproofAI support?▼

It supports twelve harnesses including Claude Code, OpenAI Codex, GitHub Copilot CLI, Cursor, OpenCode, Goose, Devin, and Factory Droid, plus custom agents through the Python SDK.

How do I monitor my production AI agents?▼

Use the fp CLI to query sessions, errors, eval scores, and token usage, or open the local dashboard to review full agent run history and audit findings.

Can I write my own safety policies?▼

Yes. Drop a JavaScript file into .failproofai/policies/ and it loads automatically, letting you allow, deny, or add instructions to any agent tool call.

Related Repositories in Software Engineering

View All in Software Engineering→
openclawopenclaw
388.4k

openclaw

Run a personal AI assistant across your devices and chat apps

Official📦 120 Skills
obraobra
280.0k

superpowers

Gives coding agents a disciplined workflow from idea to merged code

Community📦 15 Skills
facebookfacebook
248.4k

react

AI agent skills for building, testing, and porting React core

Official📦 14 Skills