Launched January 6, 2026
FieldFlow
Turn your REST API into an optimized MCP Server
FreeAPIOpen SourceAI61 impressions#680 of its week
FieldFlow turns OpenAPI-described REST endpoints into selectively filtered tools. It generates Pydantic models and FastAPI routes that forward requests to the upstream API and return only the fields the caller asks for. An optional MCP layer exposes the same functionality to Model Context Protocol clients such as Claude Desktop.
FieldFlow converts any OpenAPI-described REST API into a filtered MCP server with FastAPI routes.
- for
- Developers building LLM agents or apps that need field-level API responses.
- license
- MIT
- v0.1.0FieldFlow v0.1.06 months ago
works withFastAPIhttpxPydanticOpenAPI 3.xModel Context ProtocolClaude Desktop
Key features
6 features of FieldFlow
- OpenAPI native — Point FieldFlow at an OpenAPI 3.x spec and it auto-discovers endpoints and schemas.
- Typed Pydantic models — Generates request/response models with full validation, aliases and nested refs.
- Selective responses — Each route accepts a fields list so callers receive only the data they request.
- FastAPI gateway — Proxies calls via httpx and serves filtered results through FastAPI routes.
- Spec-first fidelity — Handles recursive schema resolution to stay aligned with the upstream API.
- Optional MCP layer — Exposes the same filtered endpoints to Model Context Protocol clients like Claude Desktop.
Use cases
- Create a GraphQL-like endpoint for LLM agents without rewriting the backend.
- Reduce token consumption by returning only needed fields to LLMs.
- Expose existing REST services to Model Context Protocol clients such as Claude Desktop.
- Generate type-safe Pydantic models for rapid API integration.
- Add field-level filtering to any third-party REST API in minutes.
FieldFlow vs alternatives
 FieldFlow |  Postman |  Insomnia |  Bruno |  Smithery | |
|---|---|---|---|---|---|
| Best for | Selective field fetching for LLM agents | API design & testing | API client & testing | Git-friendly API client | MCP server discovery |
| Pricing | Free | Subscription | Free | Free | Free |
| DevHunt upvotes | 0 | 0 | 0 | 0 | 0 |
| Launched | Jan 2026 | — | — | — | — |
- FieldFlow vs Postman: Postman is a full API design, testing and documentation platform, not a selective field proxy.
- FieldFlow vs Insomnia: Insomnia provides REST/GraphQL clients and testing, but does not generate MCP servers.
- FieldFlow vs Bruno: Bruno is an offline-first API client focused on request management, not on automatic field filtering.
- FieldFlow vs Smithery: Smithery is a registry for MCP servers, while FieldFlow creates the MCP server itself.
FieldFlow FAQ
Do I need to modify my existing API to use FieldFlow?+
No, FieldFlow works as a proxy and requires no changes to the upstream REST service.
How does FieldFlow reduce token usage for LLMs?+
Agents can request only specific fields, so unnecessary payloads are omitted, cutting token count.
What format does FieldFlow accept for API specifications?+
It accepts OpenAPI 3.x JSON or YAML files.
Can I expose the filtered endpoints to Model Context Protocol clients?+
Yes, an optional MCP layer can be generated to serve the same filtered routes to MCP clients.
What Python libraries does FieldFlow rely on?+
It uses FastAPI for routing, httpx for proxying, and Pydantic for model generation.
Summarized by DevHunt from fieldflow.dev · Oct 4, 2026. Details may change; check the official site.
.png?width=64)
