test-scope
Pass
Audited by Gen Agent Trust Hub on Oct 29, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill defines a workflow where the agent executes commands found in external documents (plans, test criteria, or rework briefs). This creates a vulnerability surface where a malicious document could inject commands into the agent's workflow.
- Ingestion points: Commands are ingested from external sources referenced as the 'Test Commands block' and 'Rework Briefs'.
- Boundary markers: The skill explicitly identifies that these documents 'can be wrong, stale or tampered with', establishing a lack of inherent trust.
- Capability inventory: The skill utilizes the 'Bash' tool and instructions allow for the execution of repository-local test, lint, type-check, and build commands.
- Sanitization: The skill mandates reading scripts before initial execution and strictly forbids actions such as calling external services, using credentials, or writing outside the working tree without explicit user consent.
Audit Metadata