skill-security-scanner — Claude Code skill: what it does and setup
skill-security-scanner audits Claude skills for command injection, unauthorized network calls, and sensitive file access before you install them.
no license 253 installs ★ 13 updated 2026-10-09
npx skills add reason-machines/security-skills --skill skill-security-scanner -a claude-codeWhat skill-security-scanner does
The utility runs static code analysis without executing suspicious scripts. It detects calls to curl or requests, access to ~/.ssh/id_rsa, read attempts on .env files, and dangerous shell operations. It flags code injection patterns such as eval, exec, and backdoor sockets.
Issues receive severity ratings: CRITICAL, WARNING, or INFO. You can output findings directly to the terminal, export a JSON report for CI pipelines, or generate a visual HTML file. By default, the scan examines both the current directory and .claude/skills/.
Running the scanner requires Python with pip or a standalone binary built via PyInstaller. If a tool needs legitimate network traffic, you can whitelist specific endpoints: domains like api.anthropic.com and github.com are allowed out of the box.
What it helps with
- detect credential access and secret leaks in skill files
- spot destructive system operations and unauthorized shell executions
- audit local .claude/skills/ folders before running untrusted tools
- generate structured JSON reports with non-zero exit codes for CI checks
Example prompts for the agent
- Scan this Claude skill for security risks before I install it
- Run a security scan on all skill files in this repository
- Check if this third-party skill contains backdoor code or network requests
How to install skill-security-scanner
1Skills CLI (Vercel)
npx skills add reason-machines/security-skills --skill skill-security-scanner -a claude-code2Manually: put the files in place
~/.claude/skills/skill-security-scanner/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/skill-security-scanner/
1Skills CLI (Vercel)
npx skills add reason-machines/security-skills --skill skill-security-scanner -a codex2Manually: put the files in place
.agents/skills/skill-security-scanner/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add reason-machines/security-skills --skill skill-security-scanner -a cursor2Manually: put the files in place
.cursor/skills/skill-security-scanner/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install reason-machines/security-skills skills/skill-security-scanner2Manually: put the files in place
.github/skills/skill-security-scanner/SKILL.md1Skills CLI (Vercel)
npx skills add reason-machines/security-skills --skill skill-security-scanner -a gemini-cli2Platform CLI
gemini skills install https://github.com/reason-machines/security-skills --path skills/skill-security-scannerManually: put the files in place
.devin/skills/skill-security-scanner/SKILL.mdThe legacy .windsurf/skills/ path still works
Skills CLI (Vercel)
npx skills add reason-machines/security-skills --skill skill-security-scanner -a opencodeClaude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: How to install skills in Claude Code: five ways
Questions about skill-security-scanner
How to install skill-security-scanner in Claude Code?
Run npx skills add reason-machines/security-skills --skill skill-security-scanner -a claude-code in your terminal, or place the file directly at ~/.claude/skills/skill-security-scanner/SKILL.md.
What does skill-security-scanner do?
It parses Claude skill definitions without executing them to flag unsafe system calls, secret access, reverse shells, and untrusted network traffic.
Can I define custom detection rules?
Yes. Pass a YAML configuration file with the --rules custom-rules.yaml option to add your own regex patterns, severity levels, and whitelists.