The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest external data from plan.md to inform its architectural recommendations, creating a surface for indirect prompt injection.
Ingestion points: Processes plan.md using the Read tool to gather context on feature scope.
Boundary markers: Absent; there are no specific delimiters or instructions to ignore embedded commands within the input file.
Capability inventory: Utilizes Read, Grep, Glob, and Write tools to perform project analysis and document generation.
Sanitization: The skill does not implement content validation or sanitization for the ingested plan data before it is integrated into the architectural output.

system-architecture