Skip to content

Add skill-guard: 9-layer security auditor for Claude Code skills#114

Open
j4rk0r wants to merge 1 commit into
Puliczek:mainfrom
j4rk0r:add-j4rk0r-claude-skills
Open

Add skill-guard: 9-layer security auditor for Claude Code skills#114
j4rk0r wants to merge 1 commit into
Puliczek:mainfrom
j4rk0r:add-j4rk0r-claude-skills

Conversation

@j4rk0r
Copy link
Copy Markdown

@j4rk0r j4rk0r commented Apr 4, 2026

Summary

Adding skill-guard to the Tools and code section.

skill-guard is a 9-layer security auditor for Claude Code skills and plugins. It analyzes skills before installation using:

  1. Permission analysis
  2. Static pattern matching
  3. LLM semantic review
  4. Bundled script inspection
  5. Data flow analysis
  6. MCP abuse vector detection
  7. Supply chain risk assessment
  8. Reputation signal checks
  9. Anti-evasion techniques

Part of the j4rk0r/claude-skills collection. MIT licensed.

This fits the "Tools and code" section as a security tool specifically designed to protect against malicious or vulnerable Claude Code skills/plugins -- complementary to MCP security tools already listed.

Placement

Added at the top of the "Tools and code" section, following the CONTRIBUTING.md guideline of "Keep NEW on TOP of section".

skill-guard is a 9-layer security auditor for Claude Code skills
and plugins. Performs static analysis, LLM semantic review, supply
chain checks, and anti-evasion detection before skill installation.

Part of j4rk0r/claude-skills collection. MIT licensed.
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant