The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from the local repository (source code and existing documentation) to update documentation files. This creates a surface where malicious instructions hidden in the code could potentially influence the agent's behavior during the documentation refinement process.\n- Ingestion points: The skill instructions command the agent to read CONTEXT.md, CONTEXT-MAP.md, existing ADR files in docs/adr/, and explore the src/ directory (SKILL.md).\n- Boundary markers: No explicit delimiters or instructions to ignore embedded commands in analyzed files are present in the skill instructions.\n- Capability inventory: The skill has the capability to modify CONTEXT.md and create new markdown files in the docs/adr/ directory (SKILL.md, ADR-FORMAT.md, CONTEXT-FORMAT.md).\n- Sanitization: No specific sanitization or verification logic is provided for content extracted from the codebase before it is used to update documentation.

grill-with-docs