skillsentry
oleh poolguy24
Coding Agents & IDEsRingkasan
OpenClaw security audit + prompt injection detector.
Deskripsi
Apa yang Dilakukan Skill Ini
Audits a local OpenClaw install for security vulnerabilities and prompt-injection indicators. Scans gateway configurations, open ports, and known injection patterns, then outputs a JSON report. Can be scheduled to run on a recurring cadence via cron.
Combines OpenClaw-specific prompt-injection detection with general vulnerability scanning in one tool, which generic security scripts don't provide.
Kapan Menggunakannya
- Auditing an OpenClaw install after adding new skills
- Detecting prompt-injection patterns in agent interactions
- Scheduling daily automated security checks via cron
- Identifying unexpected open ports on a local setup
- Reviewing security posture before sharing or deploying an install
Contoh Alur Kerja
User asks: Run a security audit on my OpenClaw install
- 1
Launch the panel server with
node scripts/panel-server.js - 2
Open the scan UI at http://localhost:8133
- 3
Run
bash scripts/audit.shto execute the audit - 4
Review JSON output for prompt-injection hits or unexpected open ports
JSON report listing detected vulnerabilities, open ports, and any prompt-injection pattern matches, stored at logs/last-report.json
Instalasi
clawhub install poolguy24/skillsentryJalankan perintah CLI-nya, atau tempel prompt ke Claude Code, Cursor, Codex, atau coding agent apa pun yang Anda pakai.