The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes local shell commands to retrieve information from the project's existing data.
Evidence: In Phase 2, the skill is instructed to run python easysdd/tools/search-yaml.py --dir easysdd/compound --query "{能力关键词}" to find relevant capability background.
Evidence: In Phase 2 (update mode), the skill performs a git log on architecture documentation modules to track implementation changes since the last review.
[INDIRECT_PROMPT_INJECTION]: The skill contains a surface for indirect prompt injection by processing untrusted external content.
Ingestion points: Phase 2 reads various external materials, including "user-provided oral materials, product ideas, user feedback, and scattered requirement descriptions from existing feature plans."
Boundary markers: The skill does not define specific delimiters or instructions to ignore potential commands within the ingested materials.
Capability inventory: The skill possesses the ability to write to the file system (within easysdd/requirements/) and execute a local search script.
Sanitization: There is no mention of sanitizing or validating the input data before it is used to draft documentation.

easysdd-requirements