The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted user input and incorporates it into a documentation file that serves as input for later automated workflow stages.
Ingestion points: User-provided responses to open-ended questions (e.g., Q6 'Other', Q10 'Other technology preferences') are captured in SKILL.md.
Boundary markers: The skill does not explicitly define sanitization or boundary markers for user-supplied descriptions before writing them to the selection document.
Capability inventory: The skill uses the Write tool to create files in the docs/ directory.
Sanitization: There is no mention of filtering or escaping user-provided text that might contain instructions targeting subsequent agent phases (e.g., 'Technical Review' or 'Program Design').

technology-selection