The Agent Skills Directory

[PROMPT_INJECTION]: The skill presents an attack surface for indirect prompt injection as it is designed to ingest untrusted external data to inform its outputs.
Ingestion points: According to SKILL.md, the agent reads discovery documents, needs analysis, and design briefs provided by the user.
Boundary markers: The skill lacks explicit delimiters or instructions to ignore potential commands within the ingested data.
Capability inventory: The skill is configured with Read, Write, Edit, Grep, and Glob tools, which allows for filesystem operations that could be targeted by injected content.
Sanitization: There is no evidence of sanitization or content validation for the external documents processed by the skill.

design-document