The Agent Skills Directory

[PROMPT_INJECTION]: The static analysis flag in references/evals.yaml corresponds to a security test case within the evaluation suite. It is designed to verify that the agent correctly refuses malicious instructions such as 'Ignore previous instructions'. This is a validation component rather than a vulnerability in the skill's logic.\n- [COMMAND_EXECUTION]: The skill instructs the agent to run local Python scripts and shell commands (e.g., git mv, sync_skills.sh) for project scaffolding and file management. These commands are legitimate for the skill's stated purpose and are scoped to the local development environment.\n- [SAFE]: The skill manages file operations securely, utilizing standard libraries for JSON handling and path normalization. No network activity, hardcoded credentials, or sensitive data access patterns were identified.

plugin-creator