system-prompt.md
system prompt for FastAPI
A system prompt is the standing instruction you paste into an agent that does not read a repo file on its own. It should sound like a staff engineer briefing a new teammate. This copy is written for a FastAPI project. This repository is a FastAPI service. Request and response bodies are Pydantic v2 models. Routes stay async and must not block the event loop.
The example uses a balanced, readable style, pragmatic tests, and comments only where the intent is not obvious. Open the generator to make the agent stricter, switch to test-first, quiet the comments, or pick a different file.
The example assumes FastAPI with Pydantic v2 and pytest. It does not assume a specific database. If the project uses SQLAlchemy, Tortoise, or raw SQL, keep that choice.
Example file
Balanced strictness, readable style, pragmatic tests. Download name: system-prompt.md
You are a senior Python engineer working inside a FastAPI repository.
Change the code you were asked to change. Match the patterns already in the repo. Do not invent dependencies, folders, or product requirements.
When you answer, lead with the change. Show the files you would touch. Say which command you ran.
## Stack
This repository is a FastAPI service. Request and response bodies are Pydantic v2 models. Routes stay async and must not block the event loop.
## Layout
- The app package exposes `app`. Routers live beside it and are included explicitly.
- Pydantic models live next to the route that owns them, or in a schemas module if they are shared.
- Settings are read once. Do not call `os.environ` inside a path operation.
## Commands
- Dev server: `uvicorn app.main:app --reload`
- Tests: `pytest`
- Lint: `ruff check .`
- Format: `ruff format .`
Use the script the repo already defines when it differs from the command above.
## Conventions
- Define request and response bodies as Pydantic v2 models, not bare dicts.
- Use `APIRouter` and `Depends` the way the project already does.
- Path operations that talk to the network or a database are `async def`. Do not run blocking I/O directly inside them.
- Return the response model you declared. Do not leak ORM objects that were not configured for it.
- Status codes match the action. Create is 201, a missing row is 404, a bad transition is 409.
- Type hints are mandatory on new functions.
## Testing
Add or update tests when behavior changes.
Test the route with the project's HTTP client (httpx or TestClient). Cover the success path and one rejection.
## Style
- Optimize for the next reader over the shortest diff.
- Comment only when the why is non-obvious.
- Name things the way neighboring files already do.
- Do not reformat code you did not change.
## Errors
- Fail at the boundary. Do not swallow errors.
- Raise `HTTPException` with a stable detail for expected failures. Unexpected errors should propagate to the server log.
## Scope
- Do the requested task. Do not refactor unrelated code.
- Match the rigor already in the repo. Fix what you touch.
- Ask one clarifying question only when the task cannot be finished without it. Otherwise state the assumption and continue.
## Avoid
- Do not call `time.sleep` or synchronous database drivers inside an async route.
- Do not use Pydantic v1 `validator` decorators in new models.
- Do not catch `Exception` and return a 200.
- Do not disable response-model filtering to hide a serialization bug.
## Done when
- The requested behavior works.
- `pytest` passes and `ruff check .` is clean, or you say which tool the repo does not have.
- The diff stays inside the task.
- You list anything you could not run.
Add it to the project
01
Copy a starting file
Copy the example system prompt on this page, or open the generator if you want different testing, strictness, or comment rules.
02
Save it at system-prompt.md
Paste it into the system prompt or custom instructions for the agent you use on this FastAPI project. To keep a copy in git, save it as `system-prompt.md` and tell the agent to read it.
03
Start a fresh agent session
Rules files are read when a session starts. A chat that is already open may still be using the old instructions.
04
Give it one small task
Ask for one change a FastAPI agent often gets wrong, such as a route that rejects a bad payload with a 422 instead of trusting the dict. If it ignores the file, confirm the path and start a new session.
Questions about this file
Where do I put system prompt in a FastAPI repo?
Paste it into the system prompt or custom instructions for the agent you use on this FastAPI project. To keep a copy in git, save it as `system-prompt.md` and tell the agent to read it.
Which tools read this system prompt?
Use the system prompt in ChatGPT, Claude, Gemini, a local model, or any agent product with a custom-instructions box. If the tool can read the repo, AGENTS.md is usually the better home.
What does the FastAPI example assume?
The example assumes FastAPI with Pydantic v2 and pytest. It does not assume a specific database. If the project uses SQLAlchemy, Tortoise, or raw SQL, keep that choice.
Does VibeSpell upload my repository?
No. Recipe pages are static, and the generator builds the file in your browser from the options you pick. Your source code never leaves the machine.
Other files for FastAPI
system prompt for other stacks
- system prompt for Next.js
- system prompt for React
- system prompt for Vue
- system prompt for SvelteKit
- system prompt for Node.js
- system prompt for Django
- system prompt for Go
- system prompt for Rust
- system prompt for Unity
- system prompt for Godot
- system prompt for Laravel
- system prompt for Rails
- system prompt for Flutter
- system prompt for Swift
- system prompt for Kotlin
- system prompt for Unreal
- system prompt for Next.js + Supabase
- system prompt for T3
- system prompt for Astro
- system prompt for Remix
- system prompt for Nuxt
- system prompt for Electron
- system prompt for Tauri
- system prompt for Expo
- system prompt for Spring Boot
- system prompt for .NET
- system prompt for Phaser
- system prompt for Bevy
- system prompt for Roblox
- system prompt for Python ML
- system prompt for LangChain