The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes shell commands for Git operations, including branch detection, fetching, pulling with rebase, stashing, and pushing, which are standard for maintaining code and document repositories.
[COMMAND_EXECUTION]: The skill includes a step to execute a local Python script (python3 scripts/update_readme_ideas_index.py) to update the project index, which is consistent with its role as a documentation and repository management tool.
[SAFE]: The skill resolves project directory paths by reading local, non-sensitive configuration markers such as ~/.config/ideas-root.txt and ~/.openclaw/ideas-root.txt.
[SAFE]: The skill ingests information from user-provided idea.md and validate.md files to construct the PRD. Ingestion points: Phase 1 and Phase 2 (SKILL.md). Boundary markers: Absent. Capability inventory: git, python3, file-write (SKILL.md). Sanitization: Absent. This data flow is central to the skill's primary function and occurs within the user's local workspace.

prd-generator