Antigravity 桌面端 & agy CLI 跨平台 Agent 安全护栏与分层规则:两档拦截强删 / 误杀代理 / 改系统代理,日志脱敏,验收先行与异构复核,194/242 项自检,一键装卸
-
Updated
Nov 15, 2026 - JavaScript
Antigravity 桌面端 & agy CLI 跨平台 Agent 安全护栏与分层规则:两档拦截强删 / 误杀代理 / 改系统代理,日志脱敏,验收先行与异构复核,194/242 项自检,一键装卸
The original 16 Claude Code safety hooks, kept for backward compatibility. The maintained set — 914 hooks, one-command install — is cc-safe-setup.
Auto-approve compound Bash commands in Claude Code — pipes, chains, subshells parsed and checked per-segment
Auto-permission engine for Antigravity CLI (agy): deterministic AST parser + LLM classifier
Guard rails for Claude Code: workspace boundaries, protected branches, production targets, lost exit status, and foreground time.
🐕 AI-Powered Risk Assessment for Claude Code — 7-layer pipeline, native notifications, menu bar dashboard. One line install.
Tool-call guardrails for Claude Code derived from cc-logger telemetry: nudge recoverable mistakes, block irreversible ones. The guard stage of callusguard; still works standalone.
Flexible and deterministic Claude Code PreToolUse handler
Claude Code-compatible command hooks for the OMP (Oh My Pi) coding agent
Moved to karlkfi/claude-bouncer. Branch-aware git permissions for Claude Code — auto-approve safe git/gh commands, pause at main and destructive ops.
The primitives layer for every coding agent — one handler API over per-agent hooks, instruction files, plugin packaging, and config.
A PreToolUse Bash tool hook that auto-approves read-only commands
An environment-aware Claude Code PreToolUse hook for the Railway CLI. Resolves the target environment at call time - from -e or the per-directory link - so production deploys, volume deletes, and live DB shells hard-deny while read-only troubleshooting stays prompt-free. Fails closed.
PreToolUse hook for Claude Code: blocks agent spending, account creation and fund moves, queues them for human approval, and proves it ran.
Stops your AI coding agent from writing calls to functions that don't exist, before the edit lands. Deterministic, no LLM, no API keys.
Block irreversible AI-agent actions before they run — deterministic, fail-closed action veto for Claude Code, Codex, Cursor, and other tool-using agents.
Zero-trust local media privacy & anti-ban physical firewall (PreToolUse Hook) for AI Coding Agents (Google Antigravity & Claude Code). Codename: Heart-Under-Blade (刃下心).
Explainable Bash permission hook and progressive allowlist builder for Claude Code
Road signs for coding agents: one measured fact at the moment of action, silence otherwise
한글 깨짐(Hangul corruption) 방지 훅 — LLM이 tool call 파라미터에 한글을 잘못 써서 「추천」이 「추첸」이 되는 문제를 저장 직전에 차단합니다. 인코딩·폰트 깨짐(mojibake)과 다른 문제. Claude Code · Cursor · Codex / Notion · Jira · Confluence
To associate your repository with the pretooluse topic, visit your repo's landing page and select "manage topics."