security-guardian
от 1999azzar
Coding Agents & IDEsКратко
Automated security auditing for OpenClaw projects.
Описание
Что делает этот навык
Scans project directories for hardcoded secrets (API keys, tokens) and Docker images for CVEs using Trivy. Outputs structured reports with file paths and line numbers. Works alongside mema-vault to move detected secrets into a credential store.
Combines secret detection and container CVE scanning in one workflow, eliminating the need to configure and interpret separate tools for each concern.
Когда его использовать
- Catching leaked API keys before a production push
- Auditing a Docker image for HIGH and CRITICAL CVEs before deployment
- Finding hardcoded tokens in source files during a code review
- Replacing plaintext credentials with vault lookups after detection
- Checking container base images for known vulnerabilities during CI
Пример сценария
User asks: 'Scan ./src for hardcoded credentials'
- 1
Run scan_secrets.py against the target directory
- 2
Parse the report for findings, noting each file path and line number
- 3
Flag each hardcoded secret as a high-severity finding
- 4
Invoke mema-vault to store the secret securely
- 5
Replace the plaintext value in source code with a vault lookup or environment variable reference
Structured report listing affected files and line numbers, with secrets removed from source and stored in mema-vault
Требования
- trivy (CLI, installed on host)
- mema-vault skill
Установка
clawhub install 1999azzar/security-guardianВыполните команду CLI или вставьте промпт в Claude Code, Cursor, Codex — любой кодинг-агент, которым пользуетесь.