Documentation Index
Fetch the complete documentation index at: https://docs.cleve.ai/llms.txt
Use this file to discover all available pages before exploring further.
Overview
Cleve exposes an MCP (Model Context Protocol) server at:
Once connected, your agent gets read/write access to the same workspace your team uses — notes, folders, skills, memory, and org context.
Authentication is handled via OAuth on the first connection. You’ll be prompted to sign in with your Cleve account; no manual token management required.
Quick Start
Claude Code
Cursor
Claude Desktop
Codex
Run this in your terminal:claude mcp add cleve --transport http https://app.cleve.ai/mcp
Add to your .cursor/mcp.json:{
"mcpServers": {
"cleve": {
"type": "http",
"url": "https://app.cleve.ai/mcp"
}
}
}
Add to your Claude Desktop configuration file:{
"mcpServers": {
"cleve": {
"type": "http",
"url": "https://app.cleve.ai/mcp"
}
}
}
Run this in your terminal:codex mcp add cleve --transport http https://app.cleve.ai/mcp
You can also find these snippets in Settings → Developer → MCP Server inside the Cleve app.
Notes
| Tool | Description | Parameters |
|---|
cleve_search_notes | Search notes by keyword | query (required), limit |
cleve_read_note | Read a note by ID | noteId (required) |
cleve_list_recent_notes | List recently updated notes | limit |
cleve_list_notes_in_folder | List notes in a specific folder | folderId (required) |
cleve_count_notes | Get total and pinned note counts | folderId |
cleve_cite | Get citation metadata for specific note IDs | noteIds[] (required) |
cleve_create_note | Create a new note | title, content, folderId, isPinned |
cleve_edit_note | Update a note’s title, content, or properties | noteId (required), title, content, folderId, isPinned |
cleve_move_note | Move a note to a different folder | noteId (required), folderId (required, null = unfiled) |
cleve_pin_note | Pin a note | noteId (required) |
cleve_unpin_note | Unpin a note | noteId (required) |
cleve_publish_note | Publish a note publicly | noteId (required) |
cleve_unpublish_note | Unpublish a note | noteId (required) |
cleve_share_note | Enable or disable link sharing for a note | noteId (required), isPublic (required) |
cleve_delete_note | Delete a note permanently | noteId (required) |
Folders
| Tool | Description | Parameters |
|---|
cleve_list_folders | List all folders | — |
cleve_create_folder | Create a new folder (optionally nested) | name (required), parentId |
cleve_rename_folder | Rename a folder | folderId (required), name (required) |
cleve_delete_folder | Delete a folder (its notes move to unfiled) | folderId (required) |
Skills Marketplace
| Tool | Description | Parameters |
|---|
cleve_list_skills | List your created and installed skills | — |
cleve_search_marketplace | Browse published skills in the marketplace | query, category |
cleve_create_skill | Create a new AI skill | name (required), prompt (required), description, category, model |
cleve_update_skill | Update an existing skill | skillId (required), name, prompt, description, category, model |
cleve_publish_skill | Publish a skill to the marketplace | skillId (required) |
cleve_unpublish_skill | Remove a skill from the marketplace | skillId (required) |
cleve_delete_skill | Delete a skill permanently | skillId (required) |
cleve_install_skill | Install a skill from the marketplace | skillId (required) |
cleve_uninstall_skill | Uninstall a skill | skillId (required) |
Memory
| Tool | Description | Parameters |
|---|
cleve_get_memory | Read the current AI memory note | — |
cleve_get_memory_settings | Read memory feature settings | — |
cleve_toggle_memory | Enable or disable memory | enabled (required) |
cleve_clear_memory | Wipe memory content and start fresh | — |
Organizations
These tools are available when you belong to one or more Cleve organizations.
| Tool | Description | Parameters |
|---|
cleve_list_orgs | List organizations you belong to (active org is marked with *) | — |
cleve_switch_org | Switch the active organization for this session | orgId (required) |
Authentication Details
The MCP server uses OAuth 2.0 via Clerk. On first connect:
- Your client opens the Cleve authorization URL
- You sign in with your Cleve account
- Your client stores the access token and sends it as a Bearer header on every request
The server reads your user identity and active organization from the token. You do not need an API key for MCP — it’s separate from the REST API key system.
If you belong to multiple organizations, use cleve_list_orgs to see them and cleve_switch_org to change context mid-session.
Example Prompts
Search my notes for anything related to "product roadmap" and summarize the key themes.
Create a folder called "2026 Planning" and move all notes with "Q1" in the title into it.
Read my memory note and tell me what context you already have about me.
Find the brand guide note and quote the tone-of-voice rules verbatim.
