wtk-qa-plan
Pass
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data which could contain malicious instructions designed to influence the agent's behavior during the planning phase.
- Ingestion points: The agent reads feature contracts, acceptance criteria, current code diffs (SKILL.md, Step 1), and repository manifests/CI workflows (references/profile.md).
- Boundary markers: There are no explicit instructions or delimiters defined to treat these external inputs as untrusted data or to ignore embedded instructions within them.
- Capability inventory: The skill is permitted to read arbitrary files within the repository and write to documentation paths under
docs/qa/(e.g., scenarios, journeys, and charters). - Sanitization: The skill lacks any defined validation or sanitization logic for the content extracted from manifests or diffs before incorporating it into the QA plan.
Audit Metadata