Setup & Installation
Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:
Version History
Latest update
What This Skill Does
Runs a structured 10-iteration security scan covering secrets exposure, OWASP basics, authentication gaps, rate limiting, and container configuration. Each iteration targets one specific check and outputs a confidence-rated result. Finishes with a summary report saved to `.ralph-report.md`.
The structured iteration loop with explicit confidence levels (VERIFIED vs PATTERN_MATCH) reduces false positives compared to unguided LLM security reviews.
When to use it
- Quick security check before pushing to production
- Catching hardcoded API keys after a refactor
- Verifying a specific security fix actually works
- Daily morning security hygiene on an active codebase
- Spot-checking a new contributor's PR for obvious issues
Example Workflow
Here's how your AI assistant might use this skill in practice.
User asks: 'Run a quick security check before I deploy'
- 1Detects stack and infra by reading package.json, Dockerfile, and .github/workflows
- 2Checks .gitignore for .env exclusion and scans for hardcoded secrets in source files
- 3Detects DEBUG mode flags and tests SQL/command injection patterns in route handlers
- 4Checks authentication middleware on sensitive endpoints and looks for rate limiting configuration
- 5Checks Dockerfile USER directive for root usage, then writes summary to .ralph-report.md
10-iteration report with PASS/FAIL/WARN per check, confidence ratings, and prioritized fix list saved to .ralph-report.md