The Agent Skills Directory

[SAFE]: The skill functions as a template for requirements engineering, describing manual processes and data extraction into static markdown files.
[PROMPT_INJECTION]: The skill processes external text inputs (project_brief, stakeholder_input), which creates a surface for indirect prompt injection. However, since the skill only outputs documentation, the risk is minimal.
Ingestion points: project_brief and stakeholder_input defined as inputs in SKILL.md.
Boundary markers: None identified in the provided file.
Capability inventory: Limited to generating markdown text (raw_requirements.md, clarifying_questions.md, etc.); no subprocess calls, eval/exec, or network operations were found in the skill content.
Sanitization: No input validation or sanitization logic is implemented.

requirements-elicitation