AI Dev Impact Lab JA
← Rankings · AI-RELATED REPO

winos1045-wq/claude-security-guard

Security hook for Claude Code to prevent destructive bash commands. Built for the Claude Builders Bounty.

GitHub ↗
★ 0
stars
75
AI relevance
50
solo dev
0
tool sigs

SUMMARY AI summary by gpt-5-mini

A small security hook designed to sit in front of Claude Code's shell execution flow to detect and block destructive bash commands. It was created for contributors to the Claude Builders Bounty. Who uses it - Developers and builders who run or allow Claude Code to execute shell commands and want a safeguard against destructive operations. What it does (key features) - Intercepts bash commands produced or forwarded by Claude Code. - Detects patterns commonly associated with destructive actions (e.g., command sequences that delete, overwrite, or corrupt data). - Blocks or sanitizes problematic commands before they are executed. - Intended as a lightweight, integrable protection layer for code-execution workflows.

DETECTED Detected AI stack

AI-related keywords found in this repo's description, topics, or README summary — grouped by category. Each badge links to the corresponding ranking detail page.

🤖 LLM models (1)
Claude

Owner

Account
winos1045-wq
Type
User
Followers
2

AI Researcher - Developer - agentic builder

Dates

Created on GitHub 2026-05-09
Last push 2026-05-09
First seen here 2026-05-09
Last fetched 2026-05-09 18:12