The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process data from the repository, including specifications, tickets, and configuration files, to clarify task requirements. This creates a surface where instructions embedded in external files could potentially influence the agent's behavior. 1. Ingestion points: Repository instructions, specs, tickets, glossary, and configuration files (defined in SKILL.md). 2. Boundary markers: Instructions explicitly require the agent to wait for user commands and respect project kits before reading files. 3. Capability inventory: The skill is authorized to write task contracts and decision records to the docs/ directory. 4. Sanitization: Explicit instructions prohibit the copying of credentials into any documented output.
[SAFE]: The repository includes a Python test script (scripts/test_engineering_core_contract.py) for static regression testing of the skill's instructions. The script reads local markdown files to ensure compliance with formatting rules and does not perform network operations or execute untrusted code.

interview-acrazie