BigHugger
GH Repository · kenryu42

cc-safety-net

A pre-execution guard for AI coding agents. It blocks destructive Git and file system commands, plus common attempts to access sensitive files, before a tool call runs. Supports Amp Code, Antigravity CLI, Claude Code, Codex, Cursor, Gemini CLI, GitHub Copilot CLI, Grok Build, Hermes Agent, Kimi Code, OpenClaw, OpenCode, and Pi.

stars
1,546
30-day movement
+62/day
Related entries
62
Connections
4
node/bunagent-appopenclawopencodeguardrailsclaude-codetypescriptantigravityclaude-code-pluginpi-extensiongemini-clicodexsecurityai-agentsai-safetycoding-agentTypeScriptdeveloper-toolsnodehermes-agentclicodex-clihooksclaude

cc-safety-net is a pre-execution guard for AI coding agents that blocks destructive Git and file system commands before tool calls run. It also blocks common attempts to access sensitive files and integrates with many agents including Claude Code, Codex, Cursor, and Gemini CLI.

It adds a guardrail layer so an agent's mistaken or malicious command is stopped before it executes.

Use it to

  • Block destructive Git commands from AI agents
  • Prevent agents from deleting or overwriting files
  • Stop access attempts to sensitive files
  • Add guardrails across multiple coding agents
  • Enforce safe tool-call execution via hooks

For Developers running AI coding agents on their machines

Role
agent-app
Language
TypeScript
Licence
MIT
Forks
78
Last push
2026-09-17
Latest release
v0.1.0 · 2025-12-25
Skills shipped
2
topicssecurityguardrailsai-agentshooksdeveloper-toolscli