clawsec-nanoclaw — Claude Code skill: vulnerability checks
clawsec-nanoclaw — a skill that monitors vulnerabilities in installed skills before adding them or during regular audits. Guards against compromised dependencies using a signed advisory feed.
AGPL-3.0 189 installs ★ 1,135 updated 2026-10-25
npx skills add prompt-security/clawsec --skill clawsec-nanoclaw -a claude-codeWhat clawsec-nanoclaw does
clawsec-nanoclaw exposes MCP tools to check skills against a curated advisory feed. The core feature is pre-installation safety checks: provide a skill name and optional version, receive a list of known issues with exploitability context. If a vulnerability is found, the skill returns remediation guidance.
Three main tools handle the workflow: pre-install checks on new skills, audits of all installed skills with severity and exploitability filtering, and browsing the full advisory feed. Additional tools verify package signatures, refresh the advisory cache every 6 hours, and check file integrity. The feed is Ed25519-signed and includes NVD CVEs, approved community advisories, and GHSA entries without CVE identifiers. Cache lives in clawsec-advisory-cache.json. Installation varies by platform—each of Claude Code, Codex, Cursor, GitHub Copilot, Gemini CLI, and OpenCode has its own command.
What it helps with
- scan a new skill for vulnerabilities before installation
- audit all current skills for security issues
- detect RCE vulnerabilities in dependencies
- verify package signatures before trusting archives
- receive alerts on critical advisories automatically
Example prompts for the agent
- Check if
new-skillversion 1.0.0 is safe to install - Show me high-exploitability vulnerabilities in my installed skills
- Schedule a daily security scan that alerts on critical or high-exploitability matches
How to install clawsec-nanoclaw
1Skills CLI (Vercel)
npx skills add prompt-security/clawsec --skill clawsec-nanoclaw -a claude-code2Manually: put the files in place
~/.claude/skills/clawsec-nanoclaw/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/clawsec-nanoclaw/
1Skills CLI (Vercel)
npx skills add prompt-security/clawsec --skill clawsec-nanoclaw -a codex2Manually: put the files in place
.agents/skills/clawsec-nanoclaw/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add prompt-security/clawsec --skill clawsec-nanoclaw -a cursor2Manually: put the files in place
.cursor/skills/clawsec-nanoclaw/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install prompt-security/clawsec skills/clawsec-nanoclaw2Manually: put the files in place
.github/skills/clawsec-nanoclaw/SKILL.md1Skills CLI (Vercel)
npx skills add prompt-security/clawsec --skill clawsec-nanoclaw -a gemini-cli2Platform CLI
gemini skills install https://github.com/prompt-security/clawsec --path skills/clawsec-nanoclawManually: put the files in place
.devin/skills/clawsec-nanoclaw/SKILL.mdThe legacy .windsurf/skills/ path still works
Skills CLI (Vercel)
npx skills add prompt-security/clawsec --skill clawsec-nanoclaw -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 clawsec-nanoclaw
How do I install clawsec-nanoclaw in Claude Code?
Run npx skills add prompt-security/clawsec --skill clawsec-nanoclaw -a claude-code, then check setup instructions in ~/.claude/skills/clawsec-nanoclaw/SKILL.md.
What does clawsec-nanoclaw do?
It scans installed and candidate skills against a signed advisory feed, flags vulnerabilities with exploitability scores, and provides context to decide whether to install or take action on an existing skill.
How often does the vulnerability database update?
The feed refreshes automatically every 6 hours. You can manually trigger an update using the clawsec_refresh_cache tool.