cporter202/agentic-ai-apis

The ultimate collection of APIs for building autonomous AI agents — 2,036 production-ready APIs across Agents, AI Models, and MCP Servers. Stop wasting weeks building infrastructure. Plug these in and ship your agent today.

What it solves

Searching for the right API to power an autonomous AI system is slow and noisy. Most directories cram in every category imaginable, so builders waste hours sifting through irrelevant listings. This project solves that by offering a tightly scoped, opinionated collection of 2,162 production-ready APIs, all focused on the infrastructure needed for autonomous agents, copilots, and tool-using assistants.

How it works

The repository is organized into three main categories. Agents (580 APIs) covers execution layers, orchestration, and autonomous task handling. AI Models (1,209 APIs) covers generation, reasoning, extraction, and transformation services. MCP Servers (373 APIs) covers Model Context Protocol integrations that connect assistants to real tools, systems, and data.

To use it, you pick the layer you need first, open that category README, scan the API names and descriptions, then click through to the provider page for implementation details, pricing, and docs. This lets you build a shortlist quickly instead of digging through unrelated topics. The collection is maintained by a daily automated sync that rebuilds only the three tracked categories when upstream data actually changes.

Who it's for

This is built for developers and teams creating autonomous assistants, AI copilots, MCP toolchains, research agents, workflow engines, tool-using LLM apps, and production AI features. It suits anyone who wants a fast, curated starting point for finding APIs that handle execution, intelligence, or tool integration in an agentic stack.

Highlights

  • 2,162 production-ready APIs in total, with 580 Agents, 1,209 AI Models, and 373 MCP Servers
  • Strictly scoped to only three categories: agents, AI models, and MCP servers, with everything else removed from the tracked structure
  • Agent-native focus: everything revolves around the stack needed for autonomous AI systems
  • Faster discovery: the curated layout reduces clutter so you can shortlist APIs in minutes
  • Daily automated maintenance that syncs the catalog and only commits when upstream data changes
  • Clear navigation per category with linked directories and provider pages for details, pricing, and docs

Related

  • Project
    mario-andreschak/FLUJO

    FLUJO is an open‑source, local‑first desktop app for building, debugging, and running AI agents visually. It lets you connect LLM providers, attach tools via the Model Context Protocol, design flows with branching/loops/sub‑flows, chat with agents, schedule automations, and expose an OpenAI‑compatible API. All workspace data stays on your computer, with optional encryption and controlled network exposure.

  • Project
    TsinghuaC3I/Awesome-Memory-for-Agents

    A curated paper list with a clear taxonomy (short‑ vs long‑term, experience vs memory) and application categories such as personalization, learning from experience, and long‑horizon tasks. It includes sections for surveys, benchmarks, and products, but only the application subsections appear in the provided excerpt.

  • Project
    Finsys/hawser

    Hawser is a lightweight Go agent that lets Dockhand manage Docker hosts from anywhere. It supports Standard mode (listens for connections) and Edge mode (initiates outbound WebSocket to Dockhand), making it work across LANs, VPSs, and NAT environments.

  • Project
    java-up-up/nexus-agent

    An enterprise-grade AI agent platform that implements a complete loop from document governance and hybrid RAG retrieval to ReAct-based agentic reasoning and MCP tool integration.