dx-tdd
Pass
Audited by Gen Agent Trust Hub on Oct 11, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill invokes test execution and git commit commands as part of the TDD workflow.
- [INDIRECT_PROMPT_INJECTION]: The skill reads external and project-internal files (
plan.md,research/) to guide its actions. It mitigates risk by specifying that external content should be treated as data, not instructions. - Ingestion points: Files in
context/changes/,research/,foundation/. - Boundary markers: Instructions to treat fetched content as data.
- Capability inventory: File read/write, git operations, test execution.
- Sanitization: Summarization of external content via
dx-references. - [DYNAMIC_EXECUTION]: The skill executes code by running tests ("Red" and "Green" phases) to verify the minimal production code passes the written tests.
Audit Metadata