The Agent Skills Directory

[SAFE]: No malicious patterns detected. The skill follows security best practices by encouraging the use of environment variables and dedicated secret management tools rather than hardcoding credentials. Documentation in SKILL.md references several local scripts (scripts/install_precommit.sh, scripts/scan_and_report.py, scripts/baseline_manager.py) that are missing from the provided skill files, which prevents complete verification of their behavior.\n- [CREDENTIALS_UNSAFE]: Static analysis flagged potential hardcoded credentials in references/detection_rules.md. Investigation confirms these are regex patterns and documentation examples used by the scanning tool to identify secrets, not actual exposed credentials.\n- [EXTERNAL_DOWNLOADS]: The skill references official resources and well-known services for its operation, such as the official Gitleaks GitHub repository, GitHub Actions, and Docker Hub. These sources are considered trusted for the purpose of the skill.\n- [COMMAND_EXECUTION]: The skill provides standard CLI commands for Gitleaks and Git operations. These are appropriate for the skill's stated purpose of secret detection and repository auditing.\n- [PROMPT_INJECTION]: The skill is susceptible to Indirect Prompt Injection (Category 8) because it processes untrusted repository data. 1. Ingestion points: gitleaks detect reads external code. 2. Boundary markers: None explicitly defined in instructions. 3. Capability inventory: Includes CLI execution and shell scripts. 4. Sanitization: Relies on regex matching. The risk is considered low and expected for security analysis tools.

secrets-gitleaks