The Agent Skills Directory

[PROMPT_INJECTION]: The skill creates a surface for indirect prompt injection by ingesting untrusted markdown artifacts from the filesystem into the agent's context.
Ingestion points: openspec/project.md, proposal.md, design.md, tasks.md, tests.md, and specs/*.md are read as inputs for review.
Boundary markers: The skill does not implement specific delimiters or 'ignore' instructions for the ingested content.
Capability inventory: The agent has access to the Bash and Agent tools while processing this data.
Sanitization: No sanitization or validation of the markdown file content is performed prior to analysis.

openspec-review