The Agent Skills Directory

[SAFE]: The skill defines a structured approach to requirements engineering, including investigation workflows, question scoring, and documentation formats. All referenced tools (Glob, Grep, Read, etc.) are standard utilities for codebase analysis.
[INDIRECT_PROMPT_INJECTION]: The skill has an attack surface for indirect prompt injection as it ingests untrusted data from user prompts and file content (via the Read tool) to generate requirements. However, this is the primary purpose of the skill, and there are no signs of malicious intent or bypass attempts.

requirements-definition