One endpoint. Every agent.
AI memory integrations for every agent.
Give Claude Code, Codex, Cursor, and ChatGPT one private OceanDB store. A decision saved in one agent can be recalled by the next, without copying profiles or hiding context inside a single product.
Verified connections
- Claude Code Coding agent · CLI and editor Connect Claude Code to a durable, self-managing memory that survives sessions and remains available to your other AI agents. Available on current Claude Code releases
- Codex Coding agent · app, CLI, and editor Give Codex durable context across tasks and repositories with one remote MCP memory, OAuth authorization, and selective recall. Available in MCP-capable Codex clients
- Cursor AI code editor Connect Cursor to a durable memory for project decisions, preferences, and handoffs that remain available outside the editor. Available where remote MCP servers are enabled
- ChatGPT General AI assistant · web Connect eligible ChatGPT plans to an external, writable MCP memory that can also be used by Claude, Cursor, and Codex. Plan-dependent custom MCP support
One remote endpoint
Every client connects to the same Streamable HTTP MCP server. There is no local database process to keep running.
Browser authorization
OAuth establishes identity without putting a long-lived API key in a repository or client configuration file.
A memory that can leave
The same store can be inspected in the dashboard and exported as Markdown or SQL.
Already use an MCP-capable agent?
Point it at the OceanDB endpoint and authorize once.