test-evidence-review
Pass
Audited by Gen Agent Trust Hub on Sep 29, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests and analyzes content from story files, unit/integration tests, and manual evidence documents. This creates a potential surface for indirect prompt injection where malicious instructions embedded in these files could attempt to influence the agent's evaluation results.\n
- Ingestion points: The skill reads various project files located in
production/sprints/,production/epics/,tests/unit/,tests/integration/, andproduction/qa/evidence/usingReadandGlobtools.\n - Boundary markers: Absent. The instructions do not employ specific delimiters or 'ignore instructions' wrappers when processing external file content.\n
- Capability inventory: The skill utilizes
Read,Glob,Grep, andWritetools, with the latter used to save optional review reports to the local file system.\n - Sanitization: No explicit sanitization, filtering, or validation of the ingested text is performed before it is processed by the agent.
Audit Metadata