Data

Tools

Indexes

Applications

Experiments

Awesome

An open API service indexing awesome lists of open source software.

https://github.com/redplanethq/core

Build your digital brain which can talk to your AI apps.
https://github.com/redplanethq/core

neo4j-graph nodejs postgresql prisma redis remix tailwind-css typescript

Last synced: 1 day ago
JSON representation

Build your digital brain which can talk to your AI apps.

Awesome Lists containing this project

README 

          


  

    🌐 Language

    


      

    


  






  

    CORE logo

  



## CORE: Your Digital Brain



**It Remembers. It Thinks. It Acts**





    

        Add to Claude Code

    

    

        Add to Cursor

    

    

        Add to OpenClaw

    

    

        Deploy on Railway

    





    

        Website

    

    

        Docs

    

    

        Discord

    






---



CORE is your digital brain - a unified, always-on AI that remembers your context, works with your AI agents, and acts across your tools.



Open source. Self-hostable. Your data, your control.



---



## What Makes a Digital Brain



### It Remembers Everything



Every conversation, decision, preference, and relationship is stored as structured knowledge in a temporal knowledge graph. Not raw text but classified facts (preferences, decisions, goals, directives) that surface exactly when you need them. Ask about something from weeks ago and the answer is there, with full context.



[Learn more about Memory →](https://docs.getcore.me/concepts/memory/overview)



### It Can Take Actions



Create GitHub issues, send Slack messages, manage your calendar without leaving the conversation. Complex multi-step workflows too: ask your coding agent to check Sentry for new errors, create a GitHub issue for each one, assign the right engineer based on git blame, and post a summary to your team's Slack channel, all from a single prompt. 200+ actions across 50+ apps through one connection point.



[Learn more about Toolkit →](https://docs.getcore.me/concepts/toolkit)



### It Thinks and Works Across Systems



Your tools are siloed. Claude Code can't talk to your WhatsApp. Cursor can't check your calendar. CORE's Agent breaks these walls - it's the orchestrator that ties memory, toolkit, and your apps together.



What the CORE Agent can do:



- **Understand intent**: classifies what you need and routes to the right tools and memory

- **Spawn other agents**: spin up a Claude Code session or browser session from WhatsApp to fix a bug or research something

- **Coordinate multi-step workflows**: pull context from memory, act across multiple apps, and report back

- **Work from anywhere**: message from WhatsApp, email, Slack, or the web dashboard — same brain, same context



[Learn more about CORE Agent →](https://docs.getcore.me/concepts/meta-agent)



### It Can Proactively Act



You shouldn't have to ask for everything. CORE monitors events - new emails, GitHub alerts, calendar changes — and evaluates them against what it knows about you. A client email arrives; CORE already knows how you handle that relationship and drafts the right response. A Sentry alert fires at night; CORE spins up a Claude Code session to investigate.



[Learn more about Concepts →](https://docs.getcore.me/concepts/overview)



---



## Architecture



CORE is built on three layers that implement these four pillars:



- **Memory**: A temporal knowledge graph that stores episodes, entities, and classified facts. Every piece of information is categorized (preference, decision, directive, goal, etc.) and connected over time. This is what makes retrieval intent-driven instead of keyword-based. [Docs →](https://docs.getcore.me/concepts/memory/overview)



- **Toolkit**: A unified actions layer for any MCP-compatible agent. Connect your apps once (GitHub, Linear, Slack, Gmail, Calendar, etc.) and every connected AI tool gets access to 200+ actions through a single endpoint. [Docs →](https://docs.getcore.me/toolkit/overview)



- **CORE Agent**: The orchestrator that ties memory and toolkit together. It understands intent, searches memory, picks tools, spawns other agents, and can act proactively based on triggers and rules. [Docs →](https://docs.getcore.me/concepts/meta-agent)



---



## What You Can Do



### Use as Your Personal Assistant



Talk to CORE from WhatsApp, email, or the web dashboard. It takes actions across your apps, manages your schedule, drafts emails, all with memory of who you are and how you work.



[Get started →](https://docs.getcore.me/quickstart/chief-of-staff)



### Supercharge Your AI Agents



Give Claude Code, Cursor, and other AI tools persistent memory and actions. Your coding agent remembers project architecture, past decisions, and preferences across sessions. No more re-explaining context.



[Get started →](https://docs.getcore.me/quickstart/supercharge-ai-agents)



Claude Code (Recommended: Plugin)



```sh

npm install -g @redplanethq/corebrain

```



Then in Claude Code:



```

/plugin marketplace add redplanethq/core

/plugin install core_brain

```



Restart Claude Code and run `/mcp` to authenticate.



The plugin auto-loads your persona (preferences, rules, decisions) at every session start and ingests conversations into memory when you're done.



Claude Code (Manual MCP)



```sh

claude mcp add --transport http --scope user core-memory https://app.getcore.me/api/v1/mcp?source=Claude-Code

```



Then type `/mcp` and open core-memory for authentication.



Cursor



[![Install MCP Server](https://cursor.com/deeplink/mcp-install-dark.svg)](https://cursor.com/en/install-mcp?name=corebrain&config=eyJ1cmwiOiJodHRwczovL2FwcC5nZXRjb3JlLm1lL2FwaS92MS9tY3A/c291cmNlPWN1cnNvciJ9)



OpenClaw



```sh

openclaw plugins install @redplanethq/openclaw-corebrain

```



Set your API key via environment variable or config:



```sh

export CORE_API_KEY=your_api_key_here

```



Get your API key from [app.getcore.me](https://app.getcore.me) → Settings → API Key.



Claude Desktop



1. Copy MCP URL: `https://app.getcore.me/api/v1/mcp?source=Claude`

2. Navigate to Settings → Connectors → Add custom connector

3. Click "Connect" and grant Claude permission to access CORE



**30+ more providers** — Windsurf, VS Code, Cline, Codex, Gemini CLI, Copilot, and more. [See all setup guides →](https://docs.getcore.me/providers/overview)



### Turn AI Chats into Memory



Sync your ChatGPT and Gemini conversations into CORE via browser extension. Searchable, reusable, and available to every connected agent.



[Get started →](https://docs.getcore.me/quickstart/capture-conversations)



---



## Quick Start



### Cloud



1. **Sign up** at [app.getcore.me](https://app.getcore.me)

2. **Connect Gmail & Calendar** — CORE scans them to learn about you so your digital brain starts with real context from day one

3. **Start using** — talk to CORE Agent, connect your AI tools, or take actions in your apps



### Self-Host



**Quick Deploy**



[![Deploy on Railway](https://railway.app/button.svg)](https://railway.com/deploy/core)



**Or with Docker**



```bash

git clone https://github.com/RedPlanetHQ/core.git

cd core

# Configure OPENAI_API_KEY in core/.env

docker-compose up -d

```



[View complete self-hosting guide →](https://docs.getcore.me/self-hosting/docker)



---



## For Agent Builders



Building AI agents? Offload memory and integrations to CORE so you can focus on your agent's logic.



- **Offload memory** — Use CORE's temporal knowledge graph as your agent's long-term memory. Store conversations, retrieve context with intent-driven search, and let your agent build knowledge over time without managing your own vector DB or graph.

- **Offload integrations** — Connect apps once in CORE, and your agent gets MCP tools for all of them. No OAuth flows to build, no API maintenance, no per-integration code.

- **Build via MCP or API** — Connect your agent to CORE via MCP (single endpoint) or use the REST API directly.



**Example Projects**



- [**core-cli**](https://github.com/RedPlanetHQ/core-cli) — Task manager agent with memory and Linear/GitHub sync

- [**holo**](https://github.com/RedPlanetHQ/holo) — Turn your CORE memory into a personal website with chat



[API Reference →](https://docs.getcore.me/api-reference) · [SDK Docs →](https://docs.getcore.me)



---



## Benchmark



CORE achieves **88.24%** average accuracy on the LoCoMo benchmark across single-hop, multi-hop, open-domain, and temporal reasoning tasks.



benchmark



[View benchmark methodology and results →](https://github.com/RedPlanetHQ/core-benchmark)



---



## Security



CASA Tier 2 Certified — third-party audited to meet Google's OAuth requirements.



- Encryption: TLS 1.3 (transit) + AES-256 (rest)

- Workspace-based isolation, role-based permissions

- Your data is never used for AI model training

- Self-hosting option for full isolation



[Security Policy →](SECURITY.md) · Vulnerability Reporting: harshith@poozle.dev



---



## Documentation



- [Welcome](https://docs.getcore.me/introduction) — Introduction to CORE

- [Concepts](https://docs.getcore.me/concepts/overview) — Memory, Agent, and Toolkit explained

- [Connect](https://docs.getcore.me/access-core/overview) — Channels and AI providers

- [Toolkit](https://docs.getcore.me/toolkit/overview) — Actions and integrations

- [Open Source](https://docs.getcore.me/guides/local-setup) — Local setup, contributing, self-hosting

- [API Reference](https://docs.getcore.me/api-reference) — REST API and endpoints

- [Changelog](https://docs.getcore.me/opensource/changelog) — Product updates



---



## Support



- Discord: [Join core-support channel](https://discord.gg/YGUZcvDjUa)

- Documentation: [docs.getcore.me](https://docs.getcore.me)

- Email: manik@poozle.dev



## Contributors