AgentSkillsCN

fastapi-service-template

从模板中生成完整的FastAPI微服务

SKILL.md
--- frontmatter
name: fastapi-service-template
description: Generate complete FastAPI microservice from template
version: 1.0.0
cross-agent: claude-code,goose
token-cost: 150

Generate FastAPI Microservice Template

Creates production-ready FastAPI microservice with OpenAI Agents SDK integration, Kubernetes deployment, and pytest configuration.

When to Use

Use this skill when you need to:

  • •Create new FastAPI microservice for LearnFlow
  • •Generate Triage, Concepts, or Code Review agent service
  • •Create service with database connection and OpenAI integration
  • •Generate Kubernetes deployment manifests
  • •Setup pytest for service testing

Usage

bash
claude "Use fastapi-service-template skill to generate <service-name>"

Parameters

  • •service-name: Name of service (triage-agent, concepts-agent, code-review-agent)
  • •port: Port number (8001, 8002, 8003) - auto-assigned if not specified
  • •include-tests: Include pytest tests (default: true)
  • •include-docker: Include Dockerfile (default: true)
  • •include-k8s: Include Kubernetes manifests (default: true)

Output

Creates directory with complete service:

code
<service-name>/
├── main.py                 # FastAPI app with health check
├── agent.py               # OpenAI Agents SDK integration
├── models.py              # Pydantic models for request/response
├── config.py              # Environment configuration
├── requirements.txt       # Python dependencies
├── Dockerfile             # Docker image definition
├── deployment.yaml        # Kubernetes deployment
├── service.yaml           # Kubernetes service
└── tests/
    └── test_agent.py      # Pytest tests

Examples

Generate Triage Agent:

bash
claude "Use fastapi-service-template skill to generate triage-agent"

Generate Concepts Agent on port 8002:

bash
claude "Use fastapi-service-template skill to generate concepts-agent with port=8002"

Generate Code Review Agent without tests:

bash
claude "Use fastapi-service-template skill to generate code-review-agent with include-tests=false"

Generated Files

main.py

FastAPI application with:

  • •/health GET endpoint (service health check)
  • •/api/query POST endpoint (agent-specific)
  • •Logging and error handling
  • •OpenAI API integration
  • •Database connection

agent.py

OpenAI Agents SDK wrapper:

  • •Agent initialization
  • •Structured outputs with Pydantic
  • •Prompt engineering for specific agent type
  • •Response formatting

models.py

Pydantic data models:

  • •QueryRequest (student query input)
  • •QueryResponse (agent response)
  • •AgentConfig (configuration)

config.py

Environment configuration:

  • •OpenAI API key loading
  • •Database connection setup
  • •Service port and host
  • •Logging configuration

requirements.txt

Python dependencies:

  • •fastapi
  • •uvicorn
  • •openai
  • •psycopg2-binary
  • •pydantic
  • •pytest

Dockerfile

Multi-stage Docker build:

  • •Python 3.11 base image
  • •Dependency caching
  • •Non-root user
  • •Health check

Kubernetes Manifests

  • •Deployment with resource limits
  • •Service for discovery
  • •ConfigMap for configuration
  • •Dapr annotations (if Tier 2)

Validation

Generated service should:

  • • Run with python main.py
  • • Respond to /health GET request
  • • Accept /api/query POST requests
  • • Return structured JSON responses
  • • Connect to PostgreSQL (if configured)
  • • Tests pass: pytest tests/
  • • Docker builds: docker build -t service:latest .

See REFERENCE.md for customization and advanced options.