test-audit
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructions require the agent to process external files and system data which creates a potential surface for indirect prompt injection.
- Ingestion points: As described in
SKILL.mdunder 'Auditing existing tests', the agent reads test files, production source code,git loghistory, CI routing, and dependency source or types. - Boundary markers: The instructions do not define boundary markers or explicit isolation for the data being ingested to prevent the agent from following instructions embedded within the analyzed code.
- Capability inventory: The skill allows the agent to modify code (deleting tests and production code seams) and execute shell commands for running tests and scripts (
SKILL.mdstep 5). - Sanitization: No sanitization or validation of the ingested content is mentioned or required by the skill logic.
Audit Metadata