The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest user journeys as the starting point for code generation. While this represents a potential surface for indirect injection, the skill prescribes a controlled, multi-step workflow that requires the generation of tests before code, which inherently limits the impact of malicious inputs.
Ingestion points: User journey descriptions provided in Step 1 of the TDD workflow.
Boundary markers: Not explicitly defined in the provided markdown instructions.
Capability inventory: Execution of local testing commands (npm test) and file system writes for code implementation.
Sanitization: No explicit input sanitization is defined, relying on the underlying agent's standard safeguards.
[COMMAND_EXECUTION]: The skill utilizes standard Node.js development commands such as npm test and npm run test:coverage. These commands are standard for the intended development environment and do not involve suspicious parameters or privilege escalation.
[DATA_EXFILTRATION]: No exfiltration vectors were identified. External service references (Supabase, Redis, OpenAI) are provided as mocks for unit testing, and E2E patterns correctly reference localhost for testing purposes.
[CREDENTIALS_UNSAFE]: No hardcoded API keys, tokens, or credentials were found. All implementation examples use placeholders or mock return values.

tdd-workflow