test-driven-development
Pass
Audited by Gen Agent Trust Hub on Oct 4, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill implements a standard Test-Driven Development workflow with quality gates and user oversight. It uses vendor-provided tools (sandbase_*) in a controlled manner, requiring explicit user approval before execution and ensuring cost transparency.
- [INDIRECT_PROMPT_INJECTION]: The skill possesses an inherent attack surface due to its core function of reading and testing local codebase files, which is necessary for TDD tasks.
- Ingestion points: The workflow involves inspecting local repository files, runtime environments, and inputs (SKILL.md).
- Boundary markers: The skill does not define specific isolation markers or warnings for instructions that might be embedded in the ingested codebase files.
- Capability inventory: The skill is authorized to execute test suites, build processes, and static analysis checks as part of its verification steps (SKILL.md).
- Sanitization: No explicit sanitization or filtering of local codebase content is mentioned before it is processed by the agent.
- Context: This surface is an expected property of a TDD utility. The risk is mitigated by instructions to maintain authorization boundaries and the requirement for user validation of all outputs.
Audit Metadata