e2e
Pass
Audited by Gen Agent Trust Hub on Aug 9, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface as it processes potentially untrusted data to guide code modifications.
- Ingestion points: Reads test output logs, error messages, and specification files (*.spec.md, SPEC.md) to categorize failures.
- Boundary markers: The instructions lack explicit delimiters or warnings to ignore instructions embedded within the processed test data.
- Capability inventory: The agent is authorized to execute shell commands via yarn and modify source code or unit tests.
- Sanitization: No specific filtering or escaping mechanisms are defined for the ingested data.
- [COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands to run tests.
- Evidence: Executes
yarn playwright test --reporter=lineandyarn test:e2eas part of the discovery and testing workflow in SKILL.md.
Audit Metadata