AI Command Center
Yet another AI agent orchestrator when n8n and Zapier already own this space.
AI-first CRM for ChatGPT, Claude and others
Plain HTTP API means agents can grep it directly without UI parsing overhead.
Developers building AI agent systems for sales or customer management
LangChain · MCP servers
Earlier this year we decided to replace a number of our internal systems with agent-first alternatives.
We couldn't find anything that worked well as a CRM. After several unsuccessful attempts to model the problem with markdown files and SQLite databases, we realised there was more to it than putting an agent on top of a database.
We didn't want to build another dashboard, pipeline view, or data-entry tool. Instead, we wanted the CRM to be something agents can operate directly to create leads, update deals, summarise accounts, track follow-ups, and generally help run the business without forcing humans to spend their day maintaining records.
The CRM is live and we use it internally.
It is open source, hackable, and designed with swappable backends, so it can run locally, on Cloudflare, Fly, AWS, or wherever you prefer.
Our own use cases are still limited, but if the project interests you, contributions are very welcome.
The goal is to rethink CRMs from the ground up, and in the process figure out what it means to build software for a different kind of user, one that does not want, or need, another data-entry application.
Feedback is more than welcome, and thank you for your time.
Yet another AI agent orchestrator when n8n and Zapier already own this space.
PostHog → CRM lead routing for account expansion—useful niche, but Vitally and Gainsight own this.
Navigation-over-search architecture sounds great until you realize it's just another RAG wrapper.
MCP-first CRM where agents and humans call the same tools through one audit log.
No SDK needed—any HTTP agent can join the conversation with just a URL and code.
Puppeteer for TUIs using libghostty-vt—finally agents can drive vim without screen scraping.