explaingit

aipotheosis-labs/aci

4,769PythonAudience · developerComplexity · 3/5LicenseSetup · moderate

TLDR

ACI.dev is an open-source platform that gives AI agents plug-and-play access to over 600 services like Slack, Google Calendar, and Vercel, handling all login flows and permissions so you can skip writing API clients from scratch.

Mindmap

mindmap
  root((ACI.dev))
    What it does
      Connects agents to 600+ tools
      Handles authentication
      Manages permissions
    Integration Methods
      MCP server
      Python SDK
      TypeScript SDK
    Use Cases
      VibeOps deployments
      Personal assistant bots
      Sales workflow automation
    Tech
      Python
      TypeScript
      OAuth
      MCP protocol
Click or tap to explore — scroll the page freely

Code map

Detail Auto

An interactive map of this repo's files and how they connect — its source is parsed live in your browser. Click Visualize to build it.

filefunction / class

Things people build with this

USE CASE 1

Build an AI coding assistant that can provision infrastructure and push code live without leaving your editor

USE CASE 2

Create a personal assistant chatbot that reads email and manages calendar events across multiple accounts

USE CASE 3

Automate sales or support workflows that need to touch Slack, a CRM, and billing tools at the same time

USE CASE 4

Connect any AI agent to 600+ services through a single MCP server instead of writing separate API clients

Tech stack

PythonTypeScriptMCPOAuth

Getting it running

Difficulty · moderate Time to first run · 1h+

Requires configuring OAuth credentials for each external service you want the agent to access.

Apache 2.0, use freely for any purpose including commercial projects, with attribution.

In plain English

ACI.dev is an open-source platform that connects AI agents to over 600 external tools and services, including things like Google Calendar, Slack, Vercel, and Supabase. Instead of each developer writing their own login flows and API clients for every service they want an AI agent to use, ACI.dev handles all of that in one place. The goal is to make it much easier to build AI agents that can take real actions in the world, not just answer questions. The platform offers two ways to plug these tools into an agent. One is a unified MCP server, which is a standardized protocol that many AI coding tools already understand. The other is a lightweight Python or TypeScript library for developers who prefer direct function calls from their own code. Both options give agents access to the same catalog of integrations and the same authentication system. Authentication is a significant part of what ACI.dev provides. When an AI agent needs to connect multiple users to multiple services, managing who can access what becomes complicated quickly. ACI.dev includes built-in support for OAuth flows, secrets storage, and permission boundaries written in plain language, so you can describe what an agent is allowed to do without writing low-level security rules. One promoted use case is what the project calls VibeOps: connecting an AI coding assistant to deployment platforms so it can handle tasks like provisioning infrastructure, configuring databases, and pushing a project live, all without the developer leaving their editor. Other uses include building personal assistant chatbots that can read email or manage calendars, and automating sales or support workflows that touch several SaaS tools at once. The entire platform, including the backend, the developer portal, and the integration definitions, is released under the Apache 2.0 license. A managed hosted version is available at aci.dev for those who do not want to run the infrastructure themselves.

Copy-paste prompts

Prompt 1
I am using ACI.dev to connect my AI agent to Google Calendar and Slack. Show me how to set up the MCP server and write a simple agent that schedules a meeting and posts a Slack notification.
Prompt 2
How do I configure ACI.dev permission boundaries so my AI agent can read emails but cannot delete or send messages without my confirmation?
Prompt 3
I want to use the ACI.dev Python SDK to build an agent that monitors Vercel deploy status and posts updates to a Slack channel. Walk me through the setup.
Prompt 4
Can I self-host ACI.dev instead of using the managed version at aci.dev? What infrastructure do I need and how do I run the backend locally?
Open on GitHub → Explain another repo

← aipotheosis-labs on gitmyhub — every repo by this author, as a profile.

Verify against the repo before relying on details.