testing-boss
Pass
Audited by Gen Agent Trust Hub on Sep 21, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill establishes workflows where the agent must process untrusted external data, which could contain malicious instructions.
- Ingestion points: In
references/ai-writes-tests.md, the agent is instructed to 'Print the full failure trace' and 'Read the SUT file referenced in the trace'. InSKILL.md, the agent is expected to review tests and triage flaky CI. - Boundary markers: The instructions lack explicit boundary markers or warnings to the agent to disregard any natural language instructions that might be embedded within the source code or test logs being analyzed.
- Capability inventory: The agent is granted capabilities to read the file system, generate new code, modify existing tests, and execute commands to run test suites.
- Sanitization: The skill does not prescribe any sanitization or validation steps for the content of failure traces or source code before the agent processes it.
Audit Metadata