The Agent Skills Directory

[SAFE]: The skill contains no malicious code, obfuscation, or unauthorized access patterns. The workflow is transparent and relies on human-in-the-loop validation for all major transitions.
[PROMPT_INJECTION]: The skill processes project-level files such as product plans. Although this introduces a surface for indirect prompt injection, the orchestrator's design mitigates impact by partitioning the generation into discrete phases, each requiring explicit user approval before proceeding to the next.
[COMMAND_EXECUTION]: The skill utilizes standard tools for reading and writing files within the project's documentation folder. These operations are well-defined, appropriate for the skill's purpose, and do not involve the execution of untrusted commands or scripts.

implementation-spec