FFT-Backend - Server-Side Implementation
Overview
Section titled “Overview”FFT-Backend is the implementation workhorse for server-side work. It masters idiomatic patterns across Node.js, Python, Ruby, Java, and Go, enforces FlowForge’s 80% coverage floor (Rule #3), and follows the correct test-location conventions for each language. For standard features it writes its own tests; for complex scenarios it delegates deeper test design to fft-testing.
Capabilities
Section titled “Capabilities”- Node.js: TypeScript-first, Fastify and NestJS patterns, ESM modules, async error handling.
- Python: FastAPI, SQLAlchemy, Pydantic v2, async/await, type-complete code with
mypy --strict. - Ruby: Rails conventions, service objects, ActiveRecord query optimization.
- Java: Spring Boot 3, Jakarta EE, reactive streams, JPA.
- Go: idiomatic concurrency, context propagation, minimal-dependency service design.
- Authentication and authorization: JWT, OAuth 2.1, RBAC, ABAC, session management.
- Data access: repository patterns, query builders, transaction boundaries, connection pooling.
- TDD execution: tests before code, correct Rule #3 test locations, 80%+ coverage enforced.
When to Use
Section titled “When to Use”- Implementing server-side logic after an API contract has been signed off by
fft-api-designer. - Porting an existing service between languages or frameworks.
- Adding authentication, background jobs, or third-party integrations.
- Refactoring to improve testability or reduce coupling.
Example Prompts
Section titled “Example Prompts”"Implement the /users endpoint in Fastify with Zod validation and repository-pattern persistence""Add OAuth 2.1 authorization-code flow to the existing Python FastAPI service""Refactor the order-processing module to use a service layer and improve unit-test coverage to 90%"