Report Abuse

Basic Information

ACI.dev is an open-source infrastructure project that provides a tool-calling platform to connect hundreds of external services to agentic IDEs and custom AI agents. The repository hosts the platform code (backend and frontend) and supports running a Unified Model-Context-Protocol (MCP) server or using lightweight SDKs so agents can invoke external tools as unified function calls. It centralizes multi-tenant authentication, OAuth flows, secrets management, dynamic tool discovery, and permission boundaries so developers do not have to build individual API clients and auth flows for each integration. The project is framework and model agnostic, released under Apache 2.0, and includes references to SDKs and a separate aci-mcp server implementation. Getting started guidance points to component READMEs for local backend and frontend development.

Links

Categorization

App Details

Features
The README highlights a number of core capabilities: more than 600 pre-built integrations to popular services, flexible access via a Unified MCP server or lightweight SDKs, and multi-tenant authentication with built-in OAuth flows and secrets management. It provides dynamic tool discovery and natural language permission boundaries to improve agent reliability. The platform logs tool usage and supports granular permissions for accountability. It is framework and model agnostic so it integrates with any LLM or agent architecture. The repo includes a backend, frontend portal, and open integration tooling. It is fully open source under Apache 2.0 and documents configuration for common hosting and platform services used in VibeOps workflows.
Use Cases
ACI.dev reduces engineering overhead by unifying authentication, discovery, and function-level access to many external services so agents can call tools reliably and securely. Developers can avoid writing separate OAuth flows and API clients for each service and instead expose capabilities as direct function calls or via a Unified MCP server. The platform improves agent accountability through natural language permission controls and tool-use logging. It accelerates VibeOps and developer workflows by enabling agentic IDEs to provision, deploy, debug, and manage infrastructure integrations with providers like Vercel, Supabase, and Cloudflare. Common use cases include personal assistant chatbots, research agents, outbound sales automation, and customer support agents that need multi-tenant, auditable access to SaaS tools.

Please fill the required fields*