The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze source code, interaction logic (state machines, reducers), and behavioral tests from an external repository to generate documentation, creating a surface for indirect prompt injection.\n
Ingestion points: The skill reads files from the target source repository during Phase 0 (reconnaissance), Phase 2 (interaction state analysis), and Phase 3 (subagent drafting).\n
Boundary markers: No specific delimiters or safety instructions (e.g., "ignore embedded instructions") are defined for handling the ingested source code content.\n
Capability inventory: The skill uses git commands (init, rev-parse), executes a local Python script (check-links.py), and mentions the use of browser tools or CLI handles for product verification in Phase 5.\n
Sanitization: No sanitization or filtering of the source code content is specified before it is processed by the agent.\n- [COMMAND_EXECUTION]: The skill requires executing command-line operations such as git init and git rev-parse to manage the documentation repository and extract project metadata.\n- [DYNAMIC_EXECUTION]: The skill includes and executes a local Python script, references/check-links.py, to perform link and anchor validation across the generated documentation files.

product-description