Stop re-explaining the repo
Cursor is strongest when it has context. Palace makes the durable part of that context searchable: conventions, fixes, preferences, commands that worked, and prior decisions.
Cursor memory
Persistent memory for Cursor
Cursor
Give Cursor a memory layer that survives sessions, remembers project decisions, and can later become shared team recall on your own infrastructure.
Works with Cursor, Claude Code, Codex, and other MCP-capable clients.
Combines semantic retrieval, BM25-style search, diaries, and knowledge graph recall.
Open-source MIT core for solo developers; paid server for shared team memory.
Palace vs built-in vendor memory
| Feature | Palace | Built-in memory |
|---|---|---|
| Cross-agent | Cursor, Claude Code, Codex, Claude Desktop | Usually one vendor product |
| MCP native | Yes | Varies; often not a shared MCP layer |
| Team memory | Shared self-hosted server | Often personal / per-account |
| Self-hosted | Yes — your infra | Typically vendor-hosted |
| Provenance | Wing / room / source citations | Often opaque personalization |
| Agent workflow | Explicit memory protocol for agents | Convenience memories inside one tool |
Beyond Cursor built-in memory
Built-in memories help inside Cursor. Palace adds a cross-agent, self-hosted layer so the same decisions follow you into Claude Code or Codex, with provenance your team can inspect.
Local first
Solo developers can start with palace-rs locally. Teams can point Cursor at a shared Palace Server when remembered context should become a team asset.
Security posture
For teams that cannot let project memory live in an opaque cloud memory layer, Palace runs inside your own network.