The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection due to the way it handles external project documentation.
Ingestion points: The skill instructions in SKILL.md direct the agent to read potentially untrusted content from docs/BRIEF.md, docs/research/*.md, and docs/screens/INDEX.md.
Boundary markers: Absent. There are no instructions providing delimiters or "ignore embedded instructions" warnings for the agent when processing these files.
Capability inventory: The skill records its output to docs/STORIES.md and invokes the ux-design tool based on the ingested content.
Sanitization: Absent. The skill does not implement validation, escaping, or filtering of the content read from the source documents before using it to influence tool calls or file writes.

user-stories