The Agent Skills Directory

[SAFE]: The skill operates entirely within the scope of project documentation and requirement analysis. It does not request or utilize tools for network communication or system-level changes.
[PROMPT_INJECTION]: The skill's instructions focus on structured workflow and methodology. No malicious directives to bypass agent guardrails or extract system prompts were identified.
[EXTERNAL_DOWNLOADS]: The skill does not reference or attempt to download any external resources, scripts, or binary files.
[REMOTE_CODE_EXECUTION]: No commands for executing code, spawning sub-processes with user input, or dynamic library loading were found in the skill or its references.
[DATA_EXFILTRATION]: There is no evidence of code that reads sensitive data or transmits information to external servers. All file operations are confined to the local project workspace.

requirements-workshop