tdd-workflow
Pass
Audited by Gen Agent Trust Hub on Oct 7, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill defines a workflow that ingests external data (feature requirements, bug reports, and planning documents from other agents) to generate code and test cases. This creates an attack surface where instructions embedded in the input data could potentially influence the agent's behavior during code generation.
- Ingestion points: Processes feature breakdowns from
dev-plannerand bug reproduction requirements. - Boundary markers: None explicitly defined to delimit user input from agent instructions.
- Capability inventory: Writing implementation and test files; executing shell scripts (
scripts/run-tests.sh,scripts/init-test.sh). - Sanitization: No specific sanitization or validation of the input requirements is mentioned.
- [DYNAMIC_EXECUTION]: The skill's primary purpose is to have the agent write implementation and test code (e.g., in Python, JS) and then execute that code to verify functionality. This follows the standard 'Script generation + execution' pattern.
- Evidence: The workflow steps (RED, GREEN, VERIFY) require creating/modifying code files and running them via
scripts/run-tests.sh.
Audit Metadata