tdd-red
Pass
Audited by Gen Agent Trust Hub on Mar 3, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [SAFE]: The skill instructions describe a legitimate development process without malicious intent.
- [NO_CODE]: No scripts or external dependencies are included; the skill relies on the agent's built-in file and command execution capabilities.
- [PROMPT_INJECTION]: The instructions do not contain patterns aimed at bypassing AI safety guardrails or overriding system prompts.
- [DATA_EXFILTRATION]: No network-enabled commands or sensitive data access patterns were found.
- [SAFE]: Evaluated for indirect prompt injection risks. The surface is considered safe as it involves standard project management files used in development.
- Ingestion points: Reads PLAN.md and LANG.md to determine task progression.
- Boundary markers: Not explicitly defined in the instructions.
- Capability inventory: The skill requires local command execution for running test suites.
- Sanitization: No specific sanitization of the content from PLAN.md is required by the instructions.
Audit Metadata