About Dejaview Mcp
Dejaview Mcp connects MCP-compatible hosts to DejaView, a persistent knowledge graph for AI agents. It stores and retrieves facts about people, projects, decisions, and relationships across sessions, with cloud and local server options.
Key features
- 01Persistent graph memory across sessions
- 02Store single or multiple facts
- 03Recall entities and search by name
- 04Natural-language graph Q&A with citations
- 05View timelines and graph statistics
- 06Share or delete graph data
Use cases
- 01Carry project context between agent sessions
- 02Record decisions and relationships
- 03Look up stored facts about a person
- 04Remove outdated facts or entities
MCP support
It exposes DejaView graph operations as MCP tools for compatible hosts, including memory storage, retrieval, search, and deletion. It can connect through a cloud endpoint or run locally.
Configuration
Remote servers need no local install. Pick your client, then copy the config or install in one click.
https://api.dejaview.io/mcpclaude mcp add --transport http DejaView-MCP https://api.dejaview.io/mcp{"mcpServers":{"DejaView-MCP":{"type":"http","url":"https://api.dejaview.io/mcp"}}}Open in Cursor{"servers":{"DejaView-MCP":{"type":"http","url":"https://api.dejaview.io/mcp"}}}Open in VS Code{"mcpServers":{"DejaView-MCP":{"type":"http","url":"https://api.dejaview.io/mcp"}}}{"mcpServers":{"DejaView-MCP":{"serverUrl":"https://api.dejaview.io/mcp"}}}Run commands in a terminal; paste JSON into the client’s MCP config and restart. Servers that need sign-in will prompt on first connect.
https://api.dejaview.io/mcpAuth requiredTools
11This tool takes no parameters.
Details
From the project READMEdejaview-mcp
Persistent knowledge graph memory for AI agents — an MCP server backed by DejaView.
Connect Claude Desktop, Cursor, Windsurf, or any MCP-compatible host to a live knowledge graph. Your AI remembers people, projects, decisions, and relationships across every session.
Get your free API key at dejaview.io.
Option 1: Cloud (no install required) ⚡
The fastest way to get started — no pip install, no local process.
{
"mcpServers": {
"dejaview": {
"type": "streamable-http",
"url": "https://api.dejaview.io/mcp",
"headers": {
"Authorization": "Bearer dv_your_key_here"
}
}
}
}
Paste this into your Claude Desktop, Cursor, or Windsurf MCP config. Done.
Option 2: Local (pip install)
If you prefer to run the server locally:
pip install dejaview-mcp
Add to your config:
{
"mcpServers": {
"dejaview": {
"command": "dejaview-mcp",
"env": {
"DEJAVIEW_API_KEY": "dv_your_key_here"
}
}
}
}
Config file locations:
- macOS:
~/Library/Application Support/Claude/claude_desktop_config.json - Linux:
~/.config/claude/claude_desktop_config.json - Windows:
%APPDATA%\Claude\claude_desktop_config.json
What it does
DejaView gives your AI a persistent knowledge graph it can read from and write to across sessions. Unlike flat context windows that reset every chat, the graph grows over time.
| Tool | What it does |
|---|---|
agent_context | Load a full memory summary at session start |
remember | Store a fact (subject, predicate, object) |
remember_many | Store multiple facts at once |
recall | Get everything known about an entity |
search | Find entities by name |
ask | Natural language Q&A over the graph with citations |
timeline | See recently stored facts |
graph_stats | Entity and relationship counts |
share | Generate a public shareable link for any entity |
forget | Remove a specific fact |
forget_entity | Remove an entity and all its connections |
Example
Once connected, just chat naturally:
"Remember that Alice is the lead on Project Atlas and prefers async communication."
The agent calls remember() automatically. Next session:
"What do I know about Alice?"
The agent calls recall("Alice") and tells you everything — including things you told it months ago.
Self-hosting
Want to run your own DejaView instance? The API is open source: github.com/JakeC77/DejaView
Set DEJAVIEW_ENDPOINT to point at your instance:
DEJAVIEW_API_KEY=dv_... DEJAVIEW_ENDPOINT=https://your-instance.com dejaview-mcp
Links
- Website: dejaview.io
- API: api.dejaview.io/docs
- GitHub: github.com/JakeC77/DejaView-MCP
Comments
New to MCP?
From what the protocol is to installing your first server, these answers cover most first questions.
An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.