At A Glance
The ultimate collection of APIs for building autonomous AI agents - 2,256 production-ready APIs across Agents, AI Models, and MCP Servers.
This repository is designed to feel like a launchpad, not a junk drawer. It is tightly scoped around the API layers that matter most when you are building autonomous systems, copilots, tool-using assistants, and MCP-native workflows.
| Metric | Count |
|---|---|
| Total APIs | 2,256 |
| Categories | 3 |
| Last Updated | 2026-10-15 |
| Focus | Agentic AI infrastructure |
Agents578 APIs Execution layers, orchestration, autonomous task handling, and agent-style workflows. |
AI Models1,318 APIs Generation, reasoning, extraction, transformation, and model-powered product building blocks. |
MCP Servers360 APIs Model Context Protocol integrations that connect assistants to real tools, systems, and data. |
Start Here
- Pick the layer you need first:
Agents,AI Models, orMCP Servers. - Open that category README and scan the API names and descriptions.
- Click through to the provider page for implementation details, pricing, and docs.
- Build your shortlist fast instead of wasting hours digging through irrelevant categories.
Explore The Stack
Agents
Best for builders who need APIs that feel closer to execution and autonomy:
- agent workflows
- orchestration layers
- autonomous task completion
- assistant behavior and action loops
AI Models
Best for builders who need intelligence and generation as a reusable service layer:
- reasoning and inference
- summarization and extraction
- content generation
- analysis and transformation
MCP Servers
Best for builders who want agents to use tools through structured interfaces:
- MCP-native tool integrations
- external system connectivity
- docs, search, analytics, scheduling, and developer workflows
- assistants that need real-world actions
Built For
| Autonomous Assistants | AI Copilots | MCP Toolchains | Internal Automation |
| Research Agents | Workflow Engines | Tool-Using LLM Apps | Production AI Features |
Why This Repo Wins
- It is opinionated. This repo is not trying to be every API category on earth.
- It is agent-native. Everything revolves around the stack needed for autonomous AI systems.
- It is faster to use. The clutter is gone, so discovery is dramatically easier.
- It is better positioned. The repo name, README, and categories now all tell the same story.
Scope Guarantee
This repository intentionally includes only:
- Agents for execution, orchestration, planning, and autonomous workflows
- AI Models for inference, generation, analysis, and reasoning
- MCP Servers for exposing tools and systems to AI assistants through MCP
Anything outside those three categories has been removed from the tracked repository structure.
Star History
Maintenance Notes
- A GitHub Actions workflow now syncs the Apify catalog daily and commits only when upstream data actually changes.
- The generation scripts in settings are configured to rebuild only the three tracked categories above.
- The visual README layout is now part of the repo's default presentation, not just a temporary pass.
- API links keep the existing affiliate tracking from the upstream source data.