Power your agent
การติดตั้งจากผู้เผยแพร่ ไม่ใช่เอนด์พอยต์ของ PlugMCP
URL แบบรีโมตหรือคำสั่งบนเครื่องของคุณ? การติดตั้งสองแบบต่างกันอย่างไร
รันบน PlugMCP
A new server you create. It is not this listing's official endpoint. Use the publisher install when you want their URL and their credentials.
เปิดจากผู้เผยแพร่
Write a SQL template and a few lines of YAML; flAPI serves it as a REST endpoint and an MCP tool for AI agents — with the same parameter validators and the same cache behind both, and role-based access control on each. One static binary with DuckDB inside: Parquet, Postgres, BigQuery, S3 and 50+ more sources.
API Framework heavily relying on the power of DuckDB and DuckDB extensions. Ready to build performant and cost-efficient APIs on top of BigQuery or Snowflake for AI Agents and Data Apps
2026-07-28 revision (dual-era: modern and legacy clients) — with the Tasks extension for long-running queries, typed schemas + structured results, OAuth discovery, per-tool RBAC, shadow/dry-run, response shaping, rate limiting, and a prompt-injection hygiene scanner{{ params.X }} references on int/double/boolean/date/time/uuid/enum/email/string fields are bound as DuckDB prepared statements — SQL injection is structurally impossible for those sitesflapii project init demos stay simpleflapi pack. scp flapi-prod user@host becomes the whole deploy. Reproducible (SOURCE_DATE_EPOCH), notarisable on macOS via a reserved Mach-O segment, with a secret deny list (*.env, secrets/*, *.pem, *.key) enforced at pack time.--no-telemetry flag, FLAPI_NO_TELEMETRY env var, or flapi.yamlThe fastest way to try flAPI — no download, no Docker:
# Run the flapi server (note: "flapi" is taken on PyPI, so the package is "flapi-io")
uvx --from flapi-io flapi -c flapi.yaml
# Run the flapii CLI client (also bundled in flapi-io)
uvx --from flapi-io flapii
Or install permanently — one package gives you both commands:
pip install flapi-io # installs both "flapi" and "flapii" commands
Pre-built binaries and Docker images are also available — see below.
The easiest way to get started with flAPI is to use the pre-built docker image.
> docker pull ghcr.io/datazoode/flapi:latest
The image is pretty small and mainly contains the flAPI binary which is statically linked against DuckDB v1.5.5. Details about the docker image can be found in the Dockerfile.
Once you have downloaded the binary, you can run flAPI by executing the following command:
> docker run -it --rm -p 8080:8080 -v $(pwd)/examples/:/config ghcr.io/datazoode/flapi -c /config/flapi.yaml
The different arguments in this docker command are:
-it --rm: Run the container in interactive mode and remove it after the process has finished-p 8080:8080: Exposes port 8080 of the container to the host. This one port serves both the REST API (http://localhost:8080) and MCP (http://localhost:8080/mcp/jsonrpc)-v $(pwd)/examples/:/config: This mounts the local examples directory to the /config directory in the container, this is where the flAPI configuration file
is expected to be found.ghcr.io/datazoode/flapi: The docker image to use-c /config/flapi.yaml: This is an argument to the flAPI application which tells it to use the flapi.yaml file in the /config directory as the configuration file.MCP is enabled by default and served by the same server as the REST API, on the same port, at
/mcp/jsonrpc — there is no separate MCP port. To switch it off, set it in flapi.yaml:
mcp:
enabled: false
If everything is set up correctly, you should be able to access the API at the URL specified in the configuration file.
> curl 'http://localhost:8080/'
___
___( o)> Welcome to
\ <_. ) flAPI
`---'
Fast and Flexible API Framework
powered by DuckDB
The flAPI server creates embedded Swagger UI at which provides an overview of the available endpoints and allows you to test them. It can be found at
You should see the familiar Swagger UI page:
The raw yaml Swagger 2.0 is also available at http://localhost:8080/doc.yaml
If MCP is enabled, you can test the MCP server as well:
# Check MCP server health
> curl 'http://localhost:8080/mcp/health'
{"status":"healthy","server":"flapi-mcp-server","version":"0.3.0","protocol_version":"2024-11-05","tools_count":0}
# Initialize MCP connection
> curl -X POST http://localhost:8080/mcp/jsonrpc \
-H "Content-Type: application/json" \
-d '{"jsonrpc": "2.0", "id": 1, "method": "initialize"}'
# List available tools
> curl -X POST http://localhost:8080/mcp/jsonrpc \
-H "Content-Type: application/json" \
-d '{"jsonrpc": "2.0", "id": 2, "method": "tools/list"}'
flAPI now supports the Model Context Protocol (MCP) in a unified configuration approach. Every flAPI instance automatically runs both a REST API server and an MCP server concurrently, allowing you to create AI tools alongside your REST endpoints using the same configuration files and SQL templates.
2026-07-28 (dual-era): serves the latest stateless MCP revision (server/discover, per-request metadata, cacheable results, OAuth discovery via RFC 9728) alongside the legacy initialize/session protocol — existing clients keep working unchangedasync and slow queries return a task handle immediately instead of blocking the connection; the durable task store survives a restart, with tasks/get / tasks/cancel and per-caller isolationstructuredContent, an outputSchema is learned after first use, and failures return actionable isError results the model can self-correct fromurl-path (REST), mcp-tool (MCP tool), or mcp-resource (MCP resource)allowed-roles), shadow/dry-run (_dryRun), response shaping, per-tool rate limiting, and a tool-description hygiene scannerflapi://customers/{id}), and x-mcp-header for per-tenant edge routingSee docs/MCP_REFERENCE.md — the dual-era model and all new capabilities are documented in §11.
POST /mcp/jsonrpc - Main JSON-RPC endpoint for tool callsGET /mcp/health - Health check endpointMCP is now automatically enabled - no separate configuration needed! Every flAPI instance runs both REST API and MCP servers concurrently.
Configuration files can define multiple entity types:
# Single configuration file serves as BOTH REST endpoint AND MCP tool
url-path: /customers/ # Makes this a REST endpoint
mcp-tool: # Also makes this an MCP tool
name: get_customers
description: Retrieve customer information by ID
result-mime-type: application/json
request:
- field-name: id
field-in: query
description: Customer ID
required: false
validators:
- type: int
min: 1
max: 1000000
preventSqlInjection: true
template-source: customers.sql
connection: [customers-parquet]
rate-limit:
enabled: true
max: 100
interval: 60
auth:
enabled: true
type: basic
users:
- username: admin
password: secret
roles: [admin]
# MCP Resource example
mcp-resource:
name: customer_schema
description: Customer database schema definition
mime-type: application/json
template-source: customer-schema.sql
connection: [customers-parquet]
Once MCP is enabled, you can interact with tools using JSON-RPC 2.0:
# Check MCP server health
curl 'http://localhost:8080/mcp/health'
# Initialize MCP connection
curl -X POST http://localhost:8080/mcp/jsonrpc \
-H "Content-Type: application/json" \
-d '{"jsonrpc": "2.0", "id": 1, "method": "initialize"}'
# List available tools (discovered from unified configuration)
curl -X POST http://localhost:8080/mcp/jsonrpc \
-H "Content-Type: application/json" \
-d '{"jsonrpc": "2.0", "id": 2, "method": "tools/list"}'
# Call a tool (same SQL template used for both REST and MCP)
curl -X POST http://localhost:8080/mcp/jsonrpc \
-H "Content-Type: application/json" \
-d '{"jsonrpc": "2.0", "id": 3, "method": "tools/call", "params": {"name": "get_customers", "arguments": {"id": "123"}}}'
The guides below are task-shaped — each one gets you to a working result, then links to the reference and to the implementation notes.
The source code of flAPI is written in C++ and closely resembles the DuckDB build process. A good documentation of the build process is the GitHub action in build.yaml. In essecence a few prerequisites need to be met:
In essecence a few prerequisites need to be met:
sudo apt-get install -y build-essential cmake ninja-buildgit clone --recurse-submodules https://github.com/datazoode/flapi.gitmake releaseThe build process will download and build DuckDB v1.5.5 and install the vcpkg package manager. We depend on the following vcpkg ports:
Note: MCP support is built-in and doesn't require additional dependencies beyond what's already included.
Start at the documentation index — it routes you from what you are trying to do to the right page.
flapi.yaml and endpoint keyflAPI is listed in the official MCP Registry under the name below (this line also serves as the registry's PyPI ownership marker):
mcp-name: io.github.datazoode/flapi
flAPI sends anonymous application_start and application_stop events to help the team understand adoption. No query data, credentials, or personal information is ever sent.
Opt out (any one of these is sufficient):
# One-off via CLI flag
./flapi --no-telemetry
# Per-session via environment variable
export FLAPI_NO_TELEMETRY=1
./flapi
# Permanently via config file (flapi.yaml)
telemetry:
enabled: false
See CLI Reference and Configuration Reference for full details.
We welcome contributions. Open an issue to discuss a change, or send a pull request.
flAPI is licensed under the Business Source License (BSL) Version 1.1. The BSL is a source-available license that gives you the following permissions:
For commercial licensing — embedding flAPI in a product, offering it as a hosted service, or any redistribution that the Additional Use Grant restricts — contact [email protected].
See the LICENSE file for the full text.
If you have any questions or need help, please open an issue.
If flAPI misbehaves — an endpoint that will not serve, a cache that will not invalidate,
an auth flow that will not complete — please
open an issue. Deployments differ in ways we
cannot reproduce here, so a report with your config is the fastest path to a fix. Every
JSON error response carries a report_issue link for exactly this reason.
If it saved you time, a star on the repo helps other people find it.
On an interactive start, a small banner says the same thing once a day. Under a container
or systemd there is no terminal, so it never prints — the startup log line carries the
pointer instead. Silence both with DATAZOO_NO_BANNER=1.
API Framework heavily relying on the power of DuckDB and DuckDB extensions. Ready to build performant and cost-efficient APIs on top of BigQuery or Snowflake for AI Agents and Data Apps
เครื่องมือ
ยังไม่มีรายการเครื่องมือ เซิร์ฟเวอร์แบบรีโมตจะได้รายการหลังการตรวจสอบสถานะครั้งแรก
FAQ
คัดลอกคอนฟิก stdio สำหรับ Cursor, Claude Desktop, Claude Code หรือ VS Code ได้ในหน้านี้ คอนฟิกนี้ใช้การติดตั้งของผู้เผยแพร่ ไม่ใช่พร็อกซีของ PlugMCP
ระเบียนนี้ไม่ได้ระบุตัวแปรสภาพแวดล้อมที่จำเป็น
Flapi ทำงานแบบโลคัลผ่าน stdio
ตรวจสอบล่าสุดเมื่อ 2026-10-28 API Framework heavily relying on the power of DuckDB and DuckDB extensions. Ready to build performant and cost-efficient APIs on top of BigQuery or Snowflake for AI Agents and Data Apps