seafile-mcp

An MCP server for Seafile, designed for multi-user deployments: one server instance serves many people, and each user supplies their own Seafile API token as the MCP API key.

How authentication works

The Seafile token is the API key. There is no separate account, no registration, and no credential storage — "authenticated" simply means "Seafile accepted this token."

Two kinds of Seafile token work, and the server auto-detects which one you gave it:

Token	How to get it	Scope
Library API token (recommended)	In Seafile: library → Advanced → API Token, choose read-only or read-write	One library. Cannot delete, move, or copy — Seafile itself offers no such endpoints for these tokens.
Account token	POST /api2/auth-token/ with your username and password	Your whole account, all libraries. Needed for search, move, copy, and delete.

Prefer a read-only library token unless you specifically need more. It is the single most effective way to limit what an assistant can do on your behalf.

The server is a stateless pass-through: it stores no credentials, keeps no user database, and retains nothing between requests. Isolation between users is structural — each request carries its own token and no per-user state exists to leak.

Safety

Deployments choose a mode via SEAFILE_MCP_MODE; tools outside the active tier are never registered, so a model cannot call what it cannot see.

• read_only — no mutating tools at all.
• safe_write — default. Write, upload, create directory, rename, move, copy. Every one is reversible via Seafile's file history or an inverse operation.
• full — additionally exposes delete.

Nothing this server exposes can irreversibly destroy data: deletes go to the library trash, and overwrites create a new version in file history. The trash purge endpoint is deliberately not wrapped as a tool.

Reading PDFs

(This feature is still under development, parsing functionality will be extended)

Most files in a typical Seafile library are PDFs, so seafile_read_file extracts their text layer instead of returning raw bytes as mojibake. There is no OCR — a scanned or image-only PDF comes back with an explicit notice instead of an error or garbage — and layout, tables, and images are not preserved.

A bare call returns the whole document, except that a PDF longer than a configurable page threshold (15 pages by default) is previewed (first 2 pages only) instead, to keep a single call cheap; pass start_page/end_page (1-indexed, inclusive) to read a specific range instead. The response's notice field always states the document's true page count and whether what you got was a preview or the full text, so the two are never ambiguous.

System admins can tune or disable the threshold with SEAFILE_MCP_PDF_PREVIEW_THRESHOLD_PAGES:

SEAFILE_MCP_PDF_PREVIEW_THRESHOLD_PAGES=30    # preview only past 30 pages
SEAFILE_MCP_PDF_PREVIEW_THRESHOLD_PAGES=all   # never preview; always extract the whole document

Running it

See docs/clients.md for client configuration (LibreChat, Claude Desktop, Claude Code, generic HTTP, and local stdio) and docs/deploy.md for the Docker + Caddy deployment.

# local, single user
export SEAFILE_SERVER_URL=https://seafile.example.org
export SEAFILE_API_TOKEN=your_token
uvx seafile-mcp --transport stdio

License

MIT