The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and act upon content from the local project repository, including design documents and source code.
Ingestion points: The skill explicitly reads CONTEXT.md, CONTEXT-MAP.md, and Architectural Decision Records (ADRs) within the docs/adr/ directory, as well as general source code for validation.
Boundary markers: There are no instructions provided to the agent to treat this external project data as untrusted or to use delimiters to prevent instructions within those files from overriding the skill's logic.
Capability inventory: The skill possesses the ability to read arbitrary repository files and write to documentation files (CONTEXT.md, docs/adr/*.md).
Sanitization: No sanitization or filtering of project-sourced data is performed before it is integrated into the agent's context during the 'grilling' process.

grill-me