qe-requirements-validation
Pass
Audited by Gen Agent Trust Hub on Sep 18, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process user stories, acceptance criteria, and Jira data, which are inherently untrusted sources that could contain malicious instructions designed to influence the agent's logic.
- Ingestion points: The skill reads requirements from Jira and local files (e.g., US-123 in SKILL.md and test case prompts in the evaluation suite).
- Boundary markers: Absent. The instructions do not specify clear delimiters or tell the agent to ignore instructions embedded within the requirements text.
- Capability inventory: The skill generates Gherkin (BDD) scenarios and traceability matrices. While the current implementation focuses on text generation, the generated code snippets (step definitions) could potentially be executed in downstream CI/CD pipelines.
- Sanitization: Absent. There is no evidence of input validation or sanitization to prevent requirements from being interpreted as commands by the LLM.
Audit Metadata