Provides tools for managing Duplicati backups, including checking status, running backups, viewing logs, and controlling the scheduler.
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type @ followed by the MCP server name and your instructions, e.g., "@mcp-duplicati Check the status of all my backup jobs."
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
MCP server for Duplicati backup management. Exposes 9 tools for checking status, running backups, viewing logs, and controlling the scheduler.
With uvx (recommended):
DUPLICATI_PASSWORD=yourpassword uvx mcp-duplicatiWith Docker:
docker run -i \
-e DUPLICATI_PASSWORD=yourpassword \
-e DUPLICATI_HOST=http://10.0.0.30:8200 \
ghcr.io/aaronckj/mcp-duplicati:latestAdd to Claude Code:
claude mcp add duplicati -- uvx mcp-duplicatiThen set env vars in Claude Code MCP settings (DUPLICATI_PASSWORD, DUPLICATI_HOST).
Related MCP server: duplicacy-mcp
Variable | Required | Default | Description |
| Yes | — | Duplicati web UI password |
| No |
| Duplicati host URL |
| No |
| HTTP timeout in seconds |
Tool | Description |
| Server version and current state |
| All configured backup jobs |
| Detailed status of a specific job |
| Trigger a job to run immediately |
| Active backup/restore progress |
| Available restore points for a job |
| Pause the scheduler (optional duration in seconds) |
| Resume the scheduler |
| Recent log entries (optionally filtered by job) |
git clone https://github.com/aaronckj/mcp-duplicati
cd mcp-duplicati
uv sync --extra dev
uv run pytest -vAbort a currently running backup job.
Inputs: backup_id (required).
Get detailed status of a specific backup job including destination, settings, and schedule.
Inputs: backup_id (required).
Compact the backup destination: removes unused data blocks to reclaim storage space.
Inputs: backup_id (required).
Create a new Duplicati backup job. source_paths: comma-separated local paths to back up. destination_url: Duplicati backend URL (e.g., 'file:///mnt/backup', 's3://bucket/path'). passphrase: optional AES-256 encryption key.
Inputs: name (required), source_paths (required), destination_url (required), passphrase.
Delete a backup job configuration. Does NOT delete backup data on the destination.
Inputs: backup_id (required).
Dismiss a Duplicati notification by ID.
Inputs: notification_id (required).
Export a backup job's full configuration as JSON. Use this to save/restore job definitions or migrate to another Duplicati instance.
Inputs: backup_id (required).
Retrieve recent log entries. backup_id: optional, filters to a specific job. page_size: 1-500. page: 0-indexed page number.
Inputs: page, backup_id, page_size.
Get Duplicati server-level settings (schedule, concurrency, update channel, etc.).
Inputs: no named parameters.
List all configured backup jobs with ID, name, last run, and next run.
Inputs: no named parameters.
List all pending Duplicati notifications and alerts.
Inputs: no named parameters.
List available restore points (filesets) for a backup job.
Inputs: backup_id (required).
Pause the Duplicati scheduler. duration: optional seconds (converted to HH:MM:SS for Duplicati API).
Inputs: duration.
Get current progress of any active backup or restore operation.
Inputs: no named parameters.
Repair the local database for a backup job. Rebuilds index from destination.
Inputs: backup_id (required).
Restore files from a backup to a local directory. restore_path: destination directory on this machine. source_path: optional path filter within backup (empty = all files). restore_time: 'latest' or ISO timestamp.
Inputs: backup_id (required), restore_path (required), source_path, restore_time.
Resume the Duplicati scheduler after a pause.
Inputs: no named parameters.
Trigger a backup job to run immediately.
Inputs: backup_id (required).
Search files within a backup version. path_filter: glob pattern (e.g., '*.pdf'). restore_time: 'latest' or ISO timestamp.
Inputs: backup_id (required), path_filter, restore_time.
Get Duplicati server version and current state.
Inputs: no named parameters.
Verify backup integrity by comparing local database with actual data at the destination.
Inputs: backup_id (required).
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
Scored across 21 tools
Each tool has a distinct purpose; there is no clear overlap between any two tools. Even closely related tools like list_backups and backup_status serve different granularity levels.
Most tools follow a clear verb_noun or noun_verb pattern with snake_case (e.g., create_backup, list_versions, backup_status). A few shorter names like pause, resume, and progress are acceptable but deviate slightly.
21 tools is well-suited for a comprehensive backup management server covering lifecycle, maintenance, monitoring, and configuration. All tools serve a clear purpose without unnecessary bloat.
The tool surface covers core backup operations (create, run, abort, delete, restore), maintenance (repair, compact, verify), logs, notifications, and settings. Missing an update/backup job edit tool, but the set is still very functional.
Pingstack monitors cron jobs and scheduled tasks with heartbeat checks, and checks websites, TCP ports, DNS records, SSL certificates and domain expiry. This MCP server lets your agent create, update, pause and delete checks, build dashboards and publish them as status pages, manage notification groups, and read incidents with their causes, uptime, run history and latency. Pause, delete and unpublish stay off until you turn them on. Browser sign-in, no API key needed.
The MCP server for Azure DevOps, bringing the power of Azure DevOps directly to your agents.
Remote MCP server for supportsheep: run AI interviews and manage support content for your blog.
Remote MCP server for supportsheep: run AI interviews and manage support content for your blog.
