Knowen - accelerating discovery
Knowen is an experiment in progress that is being designed to organize, preserve, and expand human knowledge - placing it in a hierarchical knowledge base (graph) that makes it readily available to learners of all backgrounds and ages.
https://knowen.org/Opens ChatGPT on the web or desktop and asks it to use the WebMCP tools available here.
Connect straight to this server’s public endpoint.
https://mcp.knowen.org/mcpWe add this server to your workspace, then open Studio — saved access, one connection to many servers, with a history of what ran.
Last probed Sep 14, 2026 · mcp.knowen.org
36tools discovered
Showing 25 of 36 from the live probe.
List Nodes
List nodes from the Knowen API.
Get Node
Get a single node by ID.
Create Node
Create a new node.
Update Node
Update an existing node. Only non-None fields are sent in the request (except title_of_change which is always sent).
Delete Node
Delete a node by ID.
Get Node Graph
Get the graph view for a specific node.
List Edges
List edges, optionally filtered by parent or child.
Create Edge
Create a new edge between two nodes.
Delete Edge
Delete an edge by ID.
Get Graph
Get a graph view of the knowledge graph.
Get Projects Graph
Get the projects subgraph.
Get Public Graph
Get the public subgraph.
Get Templates Graph
Get the templates subgraph.
Get Inbox Graph
Get the semantic Inbox graph. Returns the inbox scope of the simple (semantic) graph: private, parentless notes that have not been moved into a tree yet. The response carries ``data.attributes.scope == "inbox"`` and every graph node exposes an ``is_inbox`` flag (the replacement for the retired ``is_orphan`` flag). This endpoint takes no parameters. It is the only Inbox surface available to agent (assistant) principals — the ``/inbox`` routes reject their writes with 403. Returns: Parsed J
Search Nodes
Search across all accessible nodes in the knowledge graph.
Search Public Nodes
Search only public nodes in the knowledge graph.
Search Node Subtree
Search within a specific node's subtree (descendants). Searches across all descendant nodes of the given node, traversing both tree children (node_id) and graph children (edges).
Search Chunks
Search RAG chunks across all accessible nodes. Queries the embedding index for chunks whose vectors are most similar to ``query``. Results include similarity scores and optional parent-node context. When the embedding backend is unavailable the call returns a 503 unless ``fallback="ilike"`` is set.
Search Public Chunks
Search RAG chunks across public nodes only. Same endpoint semantics as :func:`search_chunks` but scoped to publicly visible nodes and usable without authentication.
Search Node Chunks
Search RAG chunks within a node subtree. Scopes the vector search to the subtree rooted at ``node_id``, traversing both tree children and graph edges. Returns 404 when the node is not accessible to the caller.
List Node Attachments
List attachments for a node via FileStorage v2 API. Note: ``page``, ``per_page``, and ``include`` are accepted for signature stability but are deprecated and ignored (v2 has no pagination).
Upload Node Attachment
Upload a file attachment to a node via FileStorage v2 API. The file bytes are obtained from one of two sources: - ``file_url``: Downloaded from the given URL. The filename is extracted from the URL path (or the ``Content-Disposition`` header) if not explicitly provided. - ``file_content``: A base64-encoded string of the file bytes. Requires ``filename`` to be explicitly provided. Note: ``attachment_type`` and ``description`` are accepted for signature stability but are deprecated and ig
Get Attachment
Get a single file entry by ID via FileStorage v2 API. Note: The ``attachment_id`` is now a Storage::Entry id. Legacy Attachment/BigAttachment ids will return 404 after the v2 cutover.
Delete Attachment
Delete (trash) a file entry by ID via FileStorage v2 API. Note: This moves the file to trash (recoverable). Purge is intentionally not exposed. The ``attachment_id`` is now a Storage::Entry id.
Get Big Attachment
Get a large file entry by ID via FileStorage v2 API. This is a thin alias for ``get_attachment`` - v2 has no size split. Note: The ``big_attachment_id`` is now a Storage::Entry id. Legacy BigAttachment ids will return 404 after the v2 cutover.
Get your MCP into directories
A working endpoint is step one. Directory coverage is the coordinated launch across ChatGPT, Claude, Cursor, the MCP Registry, and community indexes.
Directory coverage for brandsKnowen is an experiment in progress that is being designed to organize, preserve, and expand human knowledge - placing it in a hierarchical knowledge base (graph) that makes it readily available to learners of all backgrounds and ages.
Use the MCP endpoint listed on this page in your MCP client configuration. One-click install pills support Claude, Cursor, VS Code, and other hosts. Copy the remote MCP URL if your client needs a manual entry.
MCPBundles probed 36 tools on the live server. The tool list on this page reflects what was discovered at the last refresh — connect your client to see the full set available to your session.
No provider sign-in was required during MCPBundles' probe. Your client may still need MCPBundles credentials depending on how you connect.
Operate Knowen - accelerating discovery? Verify ownership to take over this directory entry.
This server appears in the MCPBundles directory. Verify you operate it to take over the listing — name, description, logo, contact email, and skill content. We email a 6-digit code to a maintainer address your server publishes in /.well-known/security.txt or /.well-known/mcpbundles.json. Free, takes about a minute.