Cloudreve MCP Server
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| CLOUDREVE_SCOPES | No | Space-separated scopes | |
| CLOUDREVE_BASE_URL | Yes | Your Cloudreve site, e.g. https://cloud.example.com | |
| CLOUDREVE_CLIENT_ID | Yes | OAuth Client ID | |
| CLOUDREVE_CACHE_ROOT | No | Local cache root | |
| CLOUDREVE_TOKEN_STORE | No | Path to tokens JSON (mode 0600) | |
| CLOUDREVE_REDIRECT_URI | No | Default http://127.0.0.1:53682/callback | |
| CLOUDREVE_AUTH_ENDPOINT | No | Authorization endpoint (default: {BASE}/session/authorize) | |
| CLOUDREVE_CLIENT_SECRET | Yes | OAuth Client Secret | |
| CLOUDREVE_DOWNLOAD_ROOT | No | Local download root | |
| CLOUDREVE_TOKEN_ENDPOINT | No | Token endpoint (default: {BASE}/api/v4/session/oauth/token) | |
| CLOUDREVE_REFRESH_ENDPOINT | No | Refresh endpoint (default: {BASE}/api/v4/session/token/refresh) | |
| CLOUDREVE_USERINFO_ENDPOINT | No | User-info endpoint (default: {BASE}/api/v4/session/oauth/userinfo) |
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {
"listChanged": true
} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| authorizeA | Start Cloudreve v4 OAuth (authorization code + PKCE). Opens a browser and waits for the local callback. Requires CLOUDREVE_CLIENT_ID and CLOUDREVE_CLIENT_SECRET. |
| auth_statusA | Show whether this MCP has a valid Cloudreve OAuth token stored locally. |
| logoutA | Clear locally stored Cloudreve OAuth tokens. |
| list_filesA | List files and folders under a Cloudreve path or File URI (default /). |
| get_file_idB | Resolve a remote path to its Cloudreve file/folder id. |
| get_file_propertiesC | Get file/folder properties by path, File URI, or id. |
| get_download_urlB | Get a temporary download URL for a remote file. |
| download_fileA | Download a remote file into the configured download root (default local app data). local_path is relative to that root. |
| create_directoryC | Create a remote directory. |
| rename_fileB | Rename a remote file or directory. |
| move_fileB | Move a remote file/folder into a destination directory. |
| copy_fileC | Copy a remote file/folder into a destination directory. |
| delete_fileA | Delete remote files/folders (soft-delete to trash by default). |
| upload_fileA | Upload a local file to Cloudreve. remote_path must include the filename (e.g. /docs/report.pdf). |
| get_source_urlB | Create/get a permanent-ish direct (source) link for a file. |
| get_share_urlC | Create a share link for a file or folder. |
| cache_listA | List files in the local MCP cache directory. |
| cache_write_textC | Write a text file into the local MCP cache. |
| cache_presign_urlB | Create a short-lived signed local_uri for a cache file (HMAC). Useful for verifying cache access. |
| cloudreve_download_to_cacheB | Download a Cloudreve file into the local MCP cache. |
| cloudreve_upload_from_cacheA | Upload a file from the local MCP cache to Cloudreve. |
| to_file_uriC | Convert a path to a Cloudreve File URI (cloudreve://my/...). |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 22 tools
Each tool targets a distinct operation: caching, OAuth, file operations, and URL handling have clear boundaries. Even similar actions like download_file and get_download_url differ in output (file vs URL).
While many tools follow verb_noun pattern, there is inconsistency with prefixes (cache_, cloudreve_) and a few outliers like 'authorize' (single verb) and 'to_file_uri' (preposition). Mix of snake_case and camelCase is not an issue, but naming styles vary.
22 tools is slightly above the typical ideal range (3-15) but still reasonable for a file storage server with caching and OAuth. The number is not excessive and each tool serves a clear purpose.
The tool set covers core file lifecycle: create (upload, mkdir), read (list, properties, download), update (rename, move, copy), delete (trash). Missing direct update of properties and batch operations, but these are minor gaps for the domain.