Press.
Facts, pictures and files for writing about Renor. Questions go to hello@renor.cloud.
In one line
Renor is one memory for all your coding agents. Tell one agent once, and every agent can recall it.
Every agent session starts from zero. Renor is a memory layer your agents share over MCP: they save what matters as you work and look it up before they answer.
Works with
Claude Code, Codex, Cursor, OpenCode, Pi, ChatGPT and Claude.ai
Connects at
Memory engine
Open source, at github.com/renorai/renor-engine
Website
renor.cloud
The founder
Renor is built by Patrick Passos, in Brazil. It starts from a problem every heavy AI user knows: each agent session begins from zero, so you explain your project, your decisions and your preferences again, in every tool.
The company’s mission is one line, make AI agents remember, and its promise to users is never repeat twice. It works like a lab: evidence outranks the founder, and a published number says how it was measured. The memory is the user’s, exportable as plain Markdown at any time. More in Culture.
Screenshots
- Memories. Everything the agents saved, as a map of things and topics, with the counts read from the workspace. PNG, 2880 × 1800
- Timeline. What each agent saved and changed, day by day. A changed fact keeps its earlier value in its history. PNG, 2880 × 1800
- Setup. The first screen after sign-up: pick the AIs you use, and one memory serves all of them. PNG, 2880 × 1800
The workspace shown is an invented example.
Film
The launch film, 60 seconds. MP4, 16:9 · MP4, 9:16
Numbers
Benchmark results and how we measure them: see Research.
Logo
The logo is the lowercase wordmark. In text, the name is written Renor.
Contact
Renor home Durable memory for AI. Your agents stop starting from zero.
Product How it works Docs Research Culture Press Blog Pricing Status
Account Get started Sign in
Legal Security Privacy Data protection Terms Refunds Support
© 2026 Renor