finalrun-test-and-fix
Pass
Audited by Gen Agent Trust Hub on Aug 19, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill serves as a legitimate coordination tool for a software testing and debugging lifecycle. It defines a structured loop for exploration, test generation, execution, and triage without malicious patterns.
- [DATA_EXPOSURE_AND_EXFILTRATION]: The instructions explicitly forbid the agent from fabricating secrets, credentials, or environment values. It requires the agent to hand off to the user for sensitive configuration, which is a security-positive practice.
- [INDIRECT_PROMPT_INJECTION]: The skill processes external data from test artifacts such as logs and screenshots to make decisions about code changes. 1. Ingestion points: device.log, runner.log, result.json, screenshots, and actions/ directory (SKILL.md). 2. Boundary markers: Not explicitly defined in the orchestrator logic. 3. Capability inventory: Ability to modify repository source code, edit YAML test files, and execute CLI commands via sibling skills. 4. Sanitization: The skill does not explicitly describe sanitization of log content or artifact data.
Audit Metadata