| FMRS | 62 / 100 · C | 84 / 100 · B | 79 / 100 · B |
| Reliability | 8 / 20 | 17 / 20 | 12 / 20 |
|---|
| Security and permissions | 14 / 20 | 16 / 20 | 18 / 20 |
|---|
| Maintenance | 14 / 20 | 16 / 20 | 15 / 20 |
|---|
| Documentation | 16 / 20 | 17 / 20 | 19 / 20 |
|---|
| Setup experience | 10 / 20 | 18 / 20 | 15 / 20 |
| Best for | - Users who want a real-time view of disk space distribution.
- Users who want AI-assisted disk cleanup and data management.
- Users needing a cross-platform (macOS, Windows, Linux) storage analyzer.
| - Local assistants that need to work inside explicitly approved directories
- Individuals and development teams that want path-based data boundaries
| - Serious Obsidian users who want AI agents reading and writing their vault
- Users wanting self-hosted, plugin-free operation with no external APIs
- Mobile/multi-device workflows accessing the vault remotely
- Security-conscious users (OAuth 2.1, atomic writes, container hardening)
|
| Not for | - Users who need permanent deletion (rather than moving to trash).
- Users who need content-level cloud scanning (metadata only).
- Users who cannot install Node.js for MCP usage.
| - Highly sensitive environments that cannot allow model access to local file contents
- Shared remote file-service use cases
| - Users unwilling to run Docker or self-host a server
- Non-Obsidian note tools (Notion, Logseq, etc.)
- Scenarios requiring only a stdio local process without an HTTP server
- Remote multi-device sync without an Obsidian Sync subscription (the remote image requires one)
|
| Required permissions | - Read file and folder metadata (size, path, etc.).
- Move files to the system trash (via OS APIs, recoverable).
- Access system trash size.
- Optionally listen on the LAN in a separate process (Fleet feature, off by default).
| - Read access to every local directory listed in the configuration
- Filesystem write access when write or move tools are enabled
| - Read/write access to the Obsidian vault folder (bind mount /vault, rw)
- Persistent /data volume (search index, OAuth token DB, logs)
- MCP_AUTH_TOKEN as Bearer token (also the JWT signing key)
- Obsidian Sync token for headless sync in remote mode
- Local download of embedding/reranker models (~45MB total), no external API calls
|
| Risks and side effects | - AI might accidentally trash important files, but deletions go to the system trash and can be manually restored.
- AI may scan folders with sensitive information, but data is processed locally and not uploaded.
- Cloud interactions are metadata-only; no file contents are downloaded.
- Offload requires sufficient space on the target drive; failure rolls back safely.
| - Sensitive files inside an allowed directory may enter model context
- Write and move tools change real files; keep scopes narrow and maintain backups
| - The server can read and write personal notes — guard MCP_AUTH_TOKEN carefully; leaking it exposes the whole vault
- Writes to real note files; despite atomic writes and protected paths, misconfiguration can alter data
- OAuth token DB lives on the /data volume; container compromise could expose valid sessions
- Remote deployments expose a public port — set PUBLIC_URL and reverse proxy correctly
- The remote image bundles proprietary obsidian-headless (not MIT-licensed); requires an active Obsidian Sync subscription
|
| Supported clients | Claude Desktop, Claude Code, Cursor | Claude Desktop, Cursor, Cline, Windsurf | Claude Code, Claude Desktop, Claude.ai, Cursor, OpenCode, MCP Inspector |
| Tools | 8 | 6 | 30 |