The Agent Skills Directory

[SAFE]: The skill's instructions are focused on structured requirement gathering and documentation management. All operations are confined to the local filesystem within the 'docs/' directory.
[NO_CODE]: This skill contains no executable code, scripts, or binaries. It relies entirely on natural language instructions for the AI agent.
[DATA_EXPOSURE]: While the skill reads project roadmap and todo files, these are standard project documentation. There is no access to sensitive system configuration files (~/.ssh, ~/.aws) or credentials.
[INDIRECT_PROMPT_INJECTION]:
Ingestion points: Processes untrusted user input describing new feature requests.
Boundary markers: Absent; the skill does not define specific delimiters for user-provided requirements.
Capability inventory: Reads and updates local Markdown files (docs/plan/*.md).
Sanitization: None specified. However, the risk is minimal as the output is restricted to task list updates in Markdown format, which does not trigger code execution.

requirement-analyst