The Agent Skills Directory

[SAFE]: The skill serves as a high-level orchestrator for other product-related tools. It coordinates tasks like requirement gathering and PRD generation.
[COMMAND_EXECUTION]: The skill executes local commands such as product-specify, product-clarify, product-implement, and product-analyze. These are part of a curated suite of internal lifecycle tools and do not involve external code or untrusted inputs.
[SAFE]: Human-in-the-loop gates are explicitly defined (e.g., gate-decisions in workflow.yml), ensuring that automated transitions require manual approval before proceeding to sensitive steps like PRD generation.
[SAFE]: File operations are restricted to the local workspace directory (.adlc/ and project root PRD.md), following standard patterns for project documentation management.

factory-product