The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes content from repository documentation and source code, and has the capability to write to the file system to update glossaries and ADRs. This creates a surface where instructions embedded in external files could influence the agent's behavior. Ingestion points: The skill reads existing repository guidance, glossaries, context maps, ADRs, and source code files as described in SKILL.md. Boundary markers: Instructions include requirements to preserve unrelated document sections and to distinguish between agreed content and unresolved proposals, providing structural limits. Capability inventory: The skill performs file read and write operations (glossaries and ADRs) to maintain project documentation. Sanitization: The instructions do not specify explicit sanitization or validation of the content being read from the repository before processing.

domain-modeling