The Agent Skills Directory

[SAFE]: The skill is a comprehensive security guide for prompt engineering and does not perform any file system, network, or code execution tasks.\n- [PROMPT_INJECTION]: While the skill mentions and analyzes prompt injection techniques (e.g., 'ignore previous instructions'), it does so solely for defensive education. It does not attempt to bypass or override the agent's safety protocols.\n- [DATA_EXFILTRATION]: No patterns of data exposure or exfiltration were found. The skill includes specific instructions to prevent the leakage of internal system prompts and credentials.\n- [REMOTE_CODE_EXECUTION]: There are no remote script downloads, package installations, or dynamic execution patterns present in the file.

injection-defense