agentic README

The project's README from github.com/transitive-bullshit/agentic. As one markdown file: README.md.

[!IMPORTANT] As of February, 2026, I'm archiving this project since it is no longer being actively developed. Agentic had a good run and explored some genuinely interesting ideas for the time, namely an HTTP <> MCP compatibility layer, a TS stdlib of AI functions, and cross-compat with all the major LLM libs + AI SDKs. But at the end of the day, it failed to gain traction, and I've moved on to other projects.

Agentic

Build Status Prettier Code Formatting

Agentic

You can think of Agentic as RapidAPI for LLM tools.

All tools listed on Agentic's marketplace have been carefully hand curated and are regularly tested with a comprehensive set of integration tests and evals. Agentic aims for quality, not quantity.

On the flip side, Agentic makes it easy to publish your own MCP servers & OpenAPI services to Agentic's MCP Gateway and instantly start charging for agentic tool use.

Key features

Getting started

TypeScript LLM SDKs

Agentic has first-class support for every major TS LLM SDK, including:

Publish your own MCP products

Agentic

Anyone can publish their own live MCP products with Agentic, but you'll need to submit your MCP to us before it can be listed on the main Agentic marketplace.

Join the community

Contributing

Agentic is proudly 100% open source.

Interested in contributing or building Agentic from scratch? See contributing.md.

agentic (transitive-bullshit/agentic)

billing
Searching Database…

18.1K ·GitHub
archived
What is this?A curated marketplace of tools for LLM agents — think RapidAPI for AI tools — plus an MCP gateway where anyone can publish an MCP server or OpenAPI service and charge for usage, with Stripe billing and semver versioning.
Do I install this locally, or is it a hosted service?It's hosted: tools run behind Agentic's MCP gateway on Cloudflare's global edge network, with customizable caching and rate limits. The open-source side is a TypeScript monorepo (Node >= 20) that includes SDKs and a CLI.
How do I start using these tools in my own project?Pick a tool from the MCP marketplace and call it from any MCP client, or use the one-line integrations for the major TS LLM SDKs: Vercel AI SDK, OpenAI, LangChain, LlamaIndex, Firebase Genkit, and Mastra. The 'Using tools' docs walk you through it.
Is the project still actively developed?No. The author archived it in February 2026 after it failed to gain traction, so don't expect further updates, and the hosted marketplace may not stick around either.
Is this for developers consuming tools, or for API owners who want to monetize?Both. Consumers get a hand-vetted catalog of agent-ready tools; publishers can expose an existing MCP server or OpenAPI service on the gateway and charge for it — though an MCP must be submitted for review before it's listed on the main marketplace.
Why are these tools better for LLMs than wrapping a regular API?Everything is built around what the README calls 'Agentic UX': tools are hand-crafted specifically for LLM tool use rather than being raw API wrappers, and every public listing is manually vetted and regularly tested with integration tests and evals. The stated philosophy is quality over quantity.
What does a tool publisher get out of the gateway?A fast hosted endpoint on Cloudflare's global edge network with customizable caching and rate limits, Stripe-based usage billing out of the box, and semver versioning — so you can charge for agentic tool use without running any infrastructure yourself.
What did the project leave behind when it was archived?The author highlights three ideas worth remembering: an HTTP MCP compatibility layer, a TypeScript stdlib of AI functions, and cross-compatibility with all the major LLM libraries and AI SDKs.
Size
16.6 MB
Stars
18.1K
Age
3y
Commits · 90 Days