Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type @ followed by the MCP server name and your instructions, e.g., "@idea-basin-mcp search my knowledge base for AI ethics"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
MCP server for Idea Basin — a semantic knowledge graph with vector search, hierarchical organization, and automatic content ingestion.
Gives Claude (and any MCP-compatible AI assistant) the ability to search, browse, and save to your personal knowledge base.
flowchart LR
subgraph client["AI Assistant"]
Claude["Claude / Any MCP Client"]
end
subgraph mcp["idea-basin-mcp"]
direction TB
T1["search_knowledge"]
T2["browse_nodes"]
T3["get_node"]
T4["save_resource"]
T5["fetch_and_save"]
end
subgraph basin["Idea Basin Server :3500"]
API["REST API"]
Embed["Embedder"]
Scraper["Scraper"]
end
subgraph storage["Storage"]
PG[("PostgreSQL + pgvector")]
end
Claude -- "stdio / JSON-RPC" --> mcp
mcp -- "HTTP" --> API
API --> Embed
API --> Scraper
API --> PGRelated MCP server: supermem
Tool | Description |
| Semantic similarity search across all knowledge chunks |
| Browse the knowledge graph tree structure (3 levels deep) |
| Inspect a node with its children and resources |
| Save a note, research, link, code, or idea to a node |
| Fetch a URL (web page, YouTube, GitHub, arXiv, PDF) and save it |
A running Idea Basin server (default: http://localhost:3500).
claude mcp add idea-basin -- node /path/to/idea-basin-mcp/build/index.jsOr with a custom Basin URL:
claude mcp add idea-basin -e IDEA_BASIN_URL=http://localhost:3500 -- node /path/to/idea-basin-mcp/build/index.jsAdd to ~/Library/Application Support/Claude/claude_desktop_config.json:
{
"mcpServers": {
"idea-basin": {
"command": "node",
"args": ["/path/to/idea-basin-mcp/build/index.js"],
"env": {
"IDEA_BASIN_URL": "http://localhost:3500"
}
}
}
}npx idea-basin-mcpEnv var | Default | Description |
|
| Base URL of your Idea Basin server |
npm install
npm run build # TypeScript → build/
npm run dev # Watch modeMIT
Browse the knowledge graph structure. Returns the tree from a starting node (or root) with 3 levels of children. Use this to understand how knowledge is organized before searching or saving.
Inputs: node_id.
Fetch a URL, extract its content, and save it as a resource in the knowledge graph. The Basin server handles scraping (web pages, YouTube, GitHub, arXiv, PDFs) and automatic ingestion.
Inputs: url (required), node_id (required), why.
Get detailed information about a specific node including its children and all resources. Use this to inspect what's already stored in a particular area of the knowledge graph.
Inputs: node_id (required).
Save a new resource (note, research, link, code, idea) to a specific node in the knowledge graph. The content will be automatically chunked and embedded for future semantic search.
Inputs: node_id (required), type (required), content (required), description, +2 more.
Search the knowledge base using semantic similarity. Returns ranked chunks with source info. Use this to find relevant information before creating new resources.
Inputs: query (required), node_id, limit.
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
Scored across 5 tools
Each tool serves a distinct purpose: navigating structure, ingesting external content, inspecting nodes, saving manual resources, and semantic search. There is no overlap in functionality.
All tools use consistent snake_case with verb_noun pattern (browse_nodes, fetch_and_save, get_node, save_resource, search_knowledge), making the API predictable.
With 5 tools, the server is well-scoped for knowledge graph management. Each tool is essential and the count is neither too few nor too many.
The set covers browsing, inspecting, adding (manual and fetched), and searching knowledge. Missing update/delete operations for nodes or resources, but these are not critical for core usage.
Personal wiki and memory layer for AI assistants. Persistent, structured memory across sessions.
Persistent memory and knowledge graph for AI assistants — keyword + vector + graph search.
Company brain for AI agents — temporal knowledge graph search, exploration, and durable memory.
Persistent knowledge graph for AI-augmented teams. Store decisions, findings, and standing rules across agent sessions with semantic search and typed connections. Includes cross-session memory, audit trail, workspace isolation, and secret detection. Built for teams running agents that need to remember. Free until launch with team tier as default, anon trial available.
