Stay up to date on the latest product updates, customer stories, and industry news from Redpoint. Opt-in and select the topics you'd like to learn more about!
Redpoint Interaction v7.x Documentation
Redpoint Interaction v7.x Documentation
Breadcrumbs

RedpointAI: Open-source AI agent platform

Overview

Your marketing team asks dozens of small questions a day: what campaigns are running now, why did this audience count change from last week, which offers are most successful. RedpointAI turns those questions into a conversation. It's a new, open-source AI agent platform that connects any large language model (LLM) provider, including OpenAI, Anthropic, and Azure OpenAI, directly to your RPI instance, so agents can look things up, reason about your setup, and get work done inside RPI itself.

What it is

RedpointAI is an open-source AI agent platform released under the Apache License 2.0. The initial iteration covers RPI capabilities only, see our What’s next section for info on what is to come.

At its core, RedpointAI does three things:

  • Gives you a chat interface for talking to agents

  • Exposes RPI's capabilities as standard tools that any Model Context Protocol (MCP) client can call

  • Routes each request to the right slice of those tools instead of handing the model your whole toolbox at once.

Bring your own LLM provider: there's no lock-in to a single model vendor.

Why you should care

Two problems show up constantly when you connect an LLM to a real platform. The model either doesn't have the tools it needs, or it has too many and wastes time (and tokens) figuring out which one to use. RedpointAI's Skill Router solves this by grouping RPI's tools, covering audiences, clients, files, folders, interactions, selection rules, auth, and admin, into domain-specific "skills." Instead of showing the model every tool on every turn, the router picks the right skill first, then hands a focused sub-agent only the tools that skill needs.

The payoff is faster responses, lower token costs, and more accurate tool selection. The model isn't guessing among numerous options; it's choosing among a handful of well-scoped skills. RedpointAI also ships with built-in domain expertise: always-on guidance for cross-cutting basics like client and tenant handling, plus a deeper expert that answers how-to, design, and strategy questions about building campaigns in RPI, covering everything from audience and selection rule logic to a full worked example of a win-back campaign. This expert guidance can provide on-demand expertise for users actively building in the UI.

In this initial iteration, most of the current RPI tool surface is read-only. The one write capability is creating folders. Agents can explore, explain, and recommend against your live instance today with a low-risk footprint while the platform matures. In future iterations, we plan to support design capabilities across RPI functionality (rules, audiences, interactions, offers, etc.).

RedpointAI runs as independent Docker containers for the web app, server, MCP server, and, in production, PostgreSQL. Each piece restarts, scales, and fails on its own, so a problem in one container doesn't take down the others.

Who it's for

  • Agent developers who already run an MCP client, such as Claude Desktop or Cursor, and want a working RPI tool surface to point it at.

  • Non-technical evaluators who want to see the whole thing running end to end, with no coding required, using nothing but Docker.

  • Contributors who want a full local dev environment to build and test against.

Getting started

The GitHub repository walks through setup for all three audiences in detail, so we won't repeat it step by step here. In short:

  • Evaluating RedpointAI? Clone the repo, drop an .env file with your LLM credentials into the project folder, and run docker compose up. The full stack comes up in containers and opens directly into the chat interface.

  • Building agents against RPI? Download the standalone MCP server binary for your platform and point your MCP client at http://localhost:3002/mcp. No clone and no build step required.

  • Contributing code? Clone the repo and run bun install && bun run dev to bring up the server, web app, and MCP server locally with hot reload.

Full prerequisites, environment variables, and troubleshooting steps live in docs/getting-started.md in the repo.

Licensing

RedpointAI is licensed under Apache License 2.0. Read it, fork it, extend it.

What’s next

  • MCP for Data Readiness Hub: Extending the MCP server framework to surface data readiness capabilities, enabling AI-assisted data preparation and quality workflows.

  • MCP for Redpoint Identity Studio: Building out agentic access to Identity Studio, bringing AI-powered identity resolution and graph capabilities into the platform.

  • Expanding Redpoint Interaction capabilities: Moving beyond read-only access to enable design-time actions, including creating, updating, and deleting rules, audiences, and interactions directly through the agentic platform.

  • Broader agentic ecosystem: As each product area reaches MCP coverage, customers gain a unified, conversational interface across the full Redpoint suite — from data readiness through identity resolution to campaign execution.