The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests and processes content from other markdown and YAML files within the project repository to perform behavioral and structural checks.
Ingestion points: The skill reads file content via Read, Glob, and Grep tools from the skills/ and qa/ directories.
Boundary markers: There are no explicit instructions to use delimiters or ignore embedded directives when evaluating the contents of external SKILL.md or test spec files.
Capability inventory: The skill possesses Read, Glob, Grep, and Write capabilities, allowing it to modify qa/catalog.yaml and create new report files in qa/results/ based on processed data.
Sanitization: The instructions do not define any filtering or sanitization steps to prevent malicious instructions inside the analyzed files from influencing the model's 'behavioral reasoning check' in Phase 2B.

skill-test