Repo Forensics
Offline security scanner for repos, AI skills, plugins, and MCP servers
Test report
- Verdict
- Tested · Works
- Score
- Tested
- Jul 17, 2026
- Environment
- Claude Code 2.x (agent harness)
- Upstream re-checked
- Aug 10, 2026 · d058363
Ran the bundled scanner live against a repo with a planted curl-pipe-to-shell exfil, a base64 exec payload, a prompt-injection SKILL.md and a committed .env: it returned 17 findings across 27 scanners, 9 CRITICAL, a deterministic BLOCK verdict and exit code 2, correctly correlating the credential-theft and exfiltration chains. The same scanner returned zero findings on a clean directory, so it earns its verdict rather than crying wolf, and works fully offline with hardened HTTPS-only threat-intel fetch.
Scored on four weighted criteria — install, triggering, output vs. baseline, docs. How scoring works
- Installs cleanly 5/5
- Triggers reliably 5/5
- Output vs. baseline 9/10
- Docs & honesty 5/5
What Repo Forensics does
Deep, offline-first security audit for git repos, AI-agent skills, and MCP servers. Detects prompt injection, credential exfiltration, obfuscated code, runtime dynamism, manifest drift, and known CVEs/CISA-KEV vulns. Triggers when the user wants to vet code or a skill before installing it, or asks 'is this repo safe?' Not for fixing vulns or pentesting.
How to install Repo Forensics
git clone https://github.com/alexgreensh/repo-forensics
mkdir -p ~/.claude/skills
cd repo-forensics && cp -r skills/repo-forensics ~/.claude/skills/repo-forensics
Skills live in ~/.claude/skills/ (global) or .claude/skills/
(per-project). Restart Claude Code after installing.
Commands — how to trigger Repo Forensics
-
/repo-forensicsOffline security scanner for repos, AI skills, plugins, and MCP servers
It also activates on plain-language prompts like these:
-
Scan this repo for prompt injection before I install it as a skill -
Check whether this MCP server has any credential exfiltration code -
Audit this plugin for supply-chain risks before we deploy it anywhere
Frequently asked questions
- Is the Repo Forensics skill free?
- Yes. The skill itself is free from alexgreensh/repo-forensics. SkillProof publishes the install command and an independent test verdict at no cost.
- Does Repo Forensics work with Claude Code?
- We tested it with Claude Code 2.x (agent harness) on Jul 17, 2026. Verdict: Tested · Works. Ran the bundled scanner live against a repo with a planted curl-pipe-to-shell exfil, a base64 exec payload, a prompt-injection SKILL.md and a committed .env: it returned 17 findings across 27 scanners, 9 CRITICAL, a deterministic BLOCK verdict and exit code 2, correctly correlating the credential-theft and exfiltration chains. The same scanner returned zero findings on a clean directory, so it earns its verdict rather than crying wolf, and works fully offline with hardened HTTPS-only threat-intel fetch.
- What is the Repo Forensics SkillProof Score?
- 9.6/10 — installs cleanly 5/5, triggers reliably 5/5, output vs. baseline 9/10, docs & honesty 5/5.
- How do I install Repo Forensics?
- Copy the install command from this page, run it in your terminal, and restart Claude Code. Skills live in ~/.claude/skills/ (global) or .claude/skills/ inside a project.
- Can I use Repo Forensics with Cursor, Copilot, Gemini CLI, Codex or other AI tools?
- The SKILL.md format is native to Claude (Claude Code, Desktop, claude.ai). The instructions inside adapt to other assistants: Cursor rules, GitHub Copilot instructions, Windsurf rules, Custom GPTs, AGENTS.md for OpenAI Codex, and GEMINI.md for Google Gemini CLI — our conversion guides cover each, and the free converter on the tools page does the wrapping for you.