The Agent Skills Directory

[SAFE]: The skill serves as a comprehensive set of guidelines for technical writing and editing. No malicious patterns, bypass attempts, or security risks were identified within the instructions or evaluation cases.- [NO_CODE]: The skill consists exclusively of markdown documentation and JSON-based evaluation data. There are no scripts, binaries, or configuration files that initiate code execution.- [CREDENTIALS_UNSAFE]: References to 'API key' and 'endpoint' appear in the guidelines and evaluation prompts only as hypothetical examples for documentation tasks (e.g., instructing the agent on how to describe configuration status to users). No actual secrets, credentials, or sensitive environment file paths are exposed or accessed.- [COMMAND_EXECUTION]: The skill contains examples of shell commands, such as virtual environment setup, within the documentation guidelines. These are intended as template content for the agent to include in produced documentation and do not represent a request for the agent to execute commands in the host environment.

product-writing