# Clawseccheck

> The genuine ClawSecCheck skill installed under its own name — must stay self-excluded from the installed-skill inventory.

- Skill: `gl0di/clawseccheck-2` (Agent Skill, multi-file: 3 files)
- Install (CLI): `npx skillmds@latest add gl0di/clawseccheck-2`
- Raw SKILL.md: https://api.skillmd.com/api/skills/gl0di/clawseccheck-2/raw
- Safety review: pending (external: skill-scanner PASS, skillspector PASS)
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Coding & Dev Tools
- Author: gl0di (https://skillmd.com/u/gl0di)
- Updated: 2026-09-22
- Page: https://skillmd.com/skills/gl0di/clawseccheck-2

---


# ClawSecCheck (genuine install — self-excluded by CONTENT)

The control case for own-name self-exclusion. This directory carries the real
package layout: a `checks/` package whose sources contain the distinctive engine
symbols (`def check_installed_skills`, `def vet_skill`, `_SKILL_CRIT`).

A security auditor necessarily ships attack signatures and red-team payloads as
*data*, so `checks/_engine.py` below embeds exactly the kind of string the audit
hunts for. If self-exclusion ever regressed to name-only — or stopped working at
all — those signatures would self-flag and this fixture would start reporting
findings. It must report none.