Stores conversation history and project files verbatim in a local, searchable memory palace and serves it to Claude Code, Codex, Cursor and other clients over MCP.
Use it now
Keep it with up to 9 other tools you're setting up, then share them all as one link.
Try this first
Ask your assistant, with MemPalace connected:
- Semantic Search
Use MemPalace to search stored drawers: ‹what you're looking for›
- Content Mining
Use MemPalace to mine project files and Claude, ChatGPT or Slack conversation exports: ‹what you need›
- Knowledge Graph
Use MemPalace to query time-bounded entity facts in a local SQLite knowledge graph: ‹what you're looking for›
Connect it over MCP
{
"mcpServers": {
"mempalace": {
"command": "docker",
"args": [
"run",
"-i",
"--rm",
"-v",
"mempalace-data:/data",
"ghcr.io/mempalace/mempalace"
]
}
}
}claude mcp add mempalace -- docker run -i --rm -v mempalace-data:/data ghcr.io/mempalace/mempalace
It runs on your computer, so web assistants such as Grok and ChatGPT can’t start it.
Its page: github.com/mempalace/mempalace
Install the CLI
pipx install mempalace
Its page: github.com/mempalace/mempalace
Call the API
Auth None needed
Its page: github.com/mempalace/mempalace
Compare
Press Compare to add it to the tray at the bottom of the window, then pick up to 2 others anywhere on the site.
- Access
- CLIMCPAPI
- Auth
- None
- Skills
- 5
Skills
Semantic Search
Searches stored drawers by meaning and returns verbatim content with similarity scores, filtered by wing, room or date.
Content Mining
Mines project files and Claude, ChatGPT or Slack conversation exports into wings and rooms for later recall.
Knowledge Graph
Adds, queries, invalidates and supersedes time-bounded entity facts in a local SQLite knowledge graph, with timelines.
Agent Diaries
Writes and reads per-agent diary entries so each specialist agent keeps its own wing and history in the palace.
Session Checkpoint
Saves a whole session in one call, deduplicating each item, filing new drawers and writing a single diary entry.