← ClaudeAtlas

skill-vetterlisted

Security-vet an Agent Skill, plugin folder or a GitHub repo of skills before installing or sharing it. Use whenever the user is about to install, copy, or enable a third-party skill; asks "is this skill safe", "check this SKILL.md", "audit this plugin"; or when a new skill, MCP helper or script bundle arrives from the internet. Runs a static scan (hidden Unicode, prompt-injection phrasing, curl-pipe-shell, credential paths, exfiltration endpoints, binaries, install hooks, ranking manipulation), pins the exact commit and file hashes, and reports PASS / REVIEW / FAIL with line numbers. Not a guarantee of safety.
Soum-Code/master-skill-hub · ★ 0 · AI & Automation · score 71
Install: claude install-skill Soum-Code/master-skill-hub
# Skill vetter Third-party skills are instructions plus scripts that run with the user's permissions. Treat them like a dependency from a stranger. ## Procedure 1. **Get the files without running anything.** For a GitHub repo, clone it into a *fresh empty directory* (`git clone --depth 1 <url> <dir>`) and record the commit: `git -C <dir> rev-parse HEAD`. Never `pip install`, `npm install`, or execute a script from it at this stage. 2. **Run the scanner on each skill folder** (the folder holding `SKILL.md`): ``` python "${CLAUDE_SKILL_DIR}/scripts/vet.py" <skill-dir> ``` Add `--json` for machine-readable output. Exit code: `0` pass, `1` review, `2` fail. 3. **Read the findings yourself.** The scanner is regex-based and misses cleverly worded injections. Open `SKILL.md` and every script it references. Look for: a description that promises one thing while the body does another; wording that keeps the user in the dark about what Claude is doing; network calls to anything other than the service the skill is about; reading `~/.ssh`, `.env`, browser profiles or `~/.claude`; writing to shell rc files, cron, or hooks. 4. **Report** in this shape: verdict, commit SHA, `tree_sha256`, the findings that matter (with file:line), and your own judgement in one or two sentences. Say plainly if you did not review a file. 5. **Install only with the user's yes**, pinned to the SHA you reviewed. If the repo changes later, re-vet: a clean skill today can be updated into a maliciou