The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data from requirement documents which could potentially contain malicious instructions.\n
Ingestion points: cleaned-requirements/index.md and CLAUDE.md (referenced in SKILL.md).\n
Boundary markers: The skill instructions do not define specific delimiters or instructions to ignore embedded commands within the analyzed documents.\n
Capability inventory: The skill is granted Read, Write, and Bash permissions (defined in SKILL.md frontmatter).\n
Sanitization: No explicit sanitization or validation of the requirement document content is performed before processing.

req-tester