The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to ingest and use information from external context files. This creates an attack surface where instructions embedded in those data files could influence the agent's behavior. \n
Ingestion points: The skill references CONTEXT.md and CONTEXT-MAP.md as sources for language and context in SKILL.md. \n
Boundary markers: The prompt lacks explicit delimiters or instructions to ignore embedded commands within the ingested content. \n
Capability inventory: The skill explicitly sets disable-model-invocation: true in SKILL.md, which limits the agent's ability to call other tools or models, reducing the potential impact of an injection. \n
Sanitization: No evidence of sanitization or validation of the external file content was found.

wait-what