Mengbi
Quality score 78Bflapi.io#agents#data#design#productivity

flAPI — Instant REST APIs & MCP Tools from SQL on DuckDB

flAPI turns SQL queries into production-ready REST APIs and MCP tools using DuckDB 1.5.2. Ship a single binary with auth, validation, and DuckLake caching.

Visit website

About this tool

flAPI is a development tool that turns SQL queries into production-ready REST APIs and MCP tools, powered by DuckDB 1.5.2. It ships as a single binary with built-in authentication, validation, and DuckLake caching.

Key Capabilities

  • SQL to REST APIs: Expose SQL queries directly as REST endpoints.
  • MCP Tool Generation: Create discoverable, authenticated, type-correct MCP tools for AI agents.
  • AI Agent Integration: Let agents query enterprise data through structured interfaces and modify records safely with validation.
  • Auth & Authorization: Restrict each tool to JWT/OIDC principals carrying named roles.
  • Error Separation: Protocol errors are separated from tool-execution errors so the model sees the latter and self-corrects.
  • Static Resource Descriptions: Agents benefit from a static resource describing result shapes up front, enabling query planning without trial-and-error.

Use Cases

  • Expose SQL queries as REST APIs
  • Query enterprise data via AI agents
  • Create CRUD APIs from SQL
  • Sync data bidirectionally
  • Build MCP tools for AI agents flAPI targets enterprise teams, AI/ML teams, and developers, and runs on Linux, macOS, and Windows. It follows a freemium pricing model.
Best for
  • Enterprise teams needing to expose SQL data as APIs
  • AI/ML teams building agent-accessible data tools
  • Developers creating CRUD APIs from SQL
  • Teams requiring role-based access control for data tools
Use cases
  • Expose SQL queries as REST APIs
  • Query enterprise data via AI agents
  • Create CRUD APIs from SQL
  • Sync data bidirectionally
  • Build MCP tools for AI agents
Strengths
  • Single binary deployment with built-in auth and validation
  • MCP tool generation for AI agent integration
  • JWT/OIDC role-based access control
  • Error separation enabling model self-correction
  • Static resource descriptions for query planning
  • DuckLake caching for performance
Limitations
  • Freemium pricing model with undisclosed free-tier limits
  • Requires DuckDB 1.5.2 as the underlying engine
  • Setup involves creating configuration and endpoints

Features & use cases

Turn SQL queries into REST APIsGenerate MCP tools for AI agentsJWT/OIDC role-based tool accessSeparate protocol and tool-execution errorsStatic resource descriptions for result shapesSingle binary with auth and validationDuckLake cachingAI agent integration for querying and modifying data

Resources

Learn how →Learn how →Learn how →Learn how →Learn how →Learn how →

Sources

Every profile keeps its sources and latest verification date.
01https://flapi.io/docs/examples/bigquery-caching

Similar tools

Row
row-bot.ai
81

Row-Bot is a local-first AI workbench that lets you run AI agents, tools, and workflows on your own computer. It features durable roles, visible goals, and coordinated work, with desktop and mobile apps.

Open details
P
Probe
probelabs.com
87

Probe is an AI code intelligence tool designed for multi-million line codebases, enabling automated code reviews, issue triage, and cross-team code intelligence.

Freemium · $40/yr

Open details
SpeakoFlow
speakoflow.com
83

SpeakoFlow is a free, open-source voice assistant that runs entirely on your own machine. It lets you dictate into any app, write or translate, or ask about your screen.

Open details
OpenProgram
openprogram.io
82

OpenProgram is an open-source runtime for building self-programming AI agents that create and refine their own workflows. It provides models, tools, memory, context, and multi-agent collaboration, and supports the Agentic Programming paradigm.

Open details