Now liveThe Skillselion MCP - thousands of ranked skills, loaded into your agent mid-task. No install.Get it →
alexgreensh avatar

Repo Forensics

  • 3 installs
  • 146 repo stars
  • Updated August 4, 2026
  • alexgreensh/repo-forensics

repo-forensics is a skill that performs security forensics on git repos, AI skills, and MCP servers, detecting prompt injection, credential theft, malicious dependencies, and known CVEs.

About

This skill performs security forensics on git repositories, AI agent skills, and MCP servers. It audits dependencies and detects prompt injection, credential theft, runtime dynamism, manifest drift, known CVEs, and actively exploited CISA KEV vulnerabilities. It runs behavioral detection rules stored as versioned JSON rule packs alongside algorithmic scanners for entropy, AST, DAST, and git forensics, and can auto-scan on package installs. It is explicitly not for fixing vulnerabilities or pentesting.

  • Deep security auditing for git repos, AI agent skills, and MCP servers
  • Detects prompt injection, credential theft, runtime dynamism, manifest drift, known CVEs, and CISA KEV vulns
  • Uses ~545 behavioral detection rules in versioned JSON rule packs plus algorithmic scanners

Repo Forensics by the numbers

  • 3 all-time installs (skills.sh)
  • Ranked #1,759 of 2,203 Security skills by installs in the Skillselion catalog
  • Data as of Aug 5, 2026 (Skillselion catalog sync)
At a glance

repo-forensics capabilities & compatibility

Capabilities
security audit · dependency scan · prompt injection detection · mcp security scan · secret detection
Use cases
security audit
Pricing
Free
npx skills add https://github.com/alexgreensh/repo-forensics --skill repo-forensics

Add your badge

Show developers this skill is listed on Skillselion. Paste this into your README.

Listed on Skillselion
Installs3
repo stars146
Last updatedAugust 4, 2026
Repositoryalexgreensh/repo-forensics

What it does

Audit a git repo, AI skill, or MCP server for prompt injection, credential theft, malicious dependencies, and known CVEs.

Who is it for?

Auditing a repo, AI agent skill, or MCP server for supply-chain and injection threats before trusting or installing it.

Skip if: Fixing vulnerabilities or pentesting, which the docs explicitly exclude.

When should I use this skill?

When you clone or install a repo, skill, or MCP server and need a security forensics scan before trusting it.

What you get

Produces findings with confidence scores and verdict tiers (BLOCK/WARN/INFO/SUPPRESSED) across many scanners.

  • security findings report (text/json/summary)
  • confidence-scored verdicts (BLOCK/WARN/INFO/SUPPRESSED)

By the numbers

  • ~545 behavioral detection patterns
  • 1,800+ pytest tests
  • 4 verdict tiers (BLOCK/WARN/INFO/SUPPRESSED)

Files

.gitignoreTextGitHub ↗
# Internal / self-learning notes — never commit
FIELD_NOTES.md

# Python artifacts
__pycache__/
*.py[cod]
*.egg-info/
dist/
build/

# OS / IDE
.DS_Store
Thumbs.db
.vscode/
.idea/
*.swp
*.swo

Related skills

FAQ

What does repo-forensics do?

It audits git repos, AI skills, and MCP servers for prompt injection, credential theft, runtime dynamism, manifest drift, known CVEs, and CISA KEV vulnerabilities.

Does it fix the vulnerabilities it finds?

No. The docs state it is not for fixing vulnerabilities or pentesting; it detects and reports.

How are its detection rules maintained?

About 545 behavioral patterns live in versioned JSON rule packs, updated via a signed daily rule-pack feed that only overlays when verified and newer.

Securityauditappsecsecrets

This week in AI coding

Five minutes, every Monday - the tools, releases and tactics for developers.

unsubscribe anytime.