MCP App Store
by KeenEthics
SimpleBackups icon

SimpleBackups MCP Server for Claude and ChatGPT

By SimpleBackups

Overview

SimpleBackups is automated backup management for databases, servers, cloud storage and SaaS apps. It covers Supabase, Neon, PlanetScale, PostgreSQL, MySQL, MongoDB, and apps like Notion, GitHub and Linear. This connector lets Claude check backup health, investigate failures and run safe operations without opening the dashboard. What you can do: - Check health across your account in one call: what ran, what failed, what’s gone stale. - Diagnose a failure from the run’s console output, so Claude can find the real cause (bad credentials, disk full, permission denied, timeout). - See day-by-day run history across all your backups for up to 30 days. - Run a backup before a deploy or migration, pause or resume a schedule, and prepare a restore. - 0 - Preparing a restore returns a short-lived download link or an rsync command you run yourself. Example prompts - “Does my Supabase database have a successful backup from today?” - “Why did the production Postgres backup fail last night?” - “Back up my Neon database now, before I run the migration.” - “Show this week’s run history for all my database backups.” - “Prepare a restore of yesterday’s Notion backup.” Safety: 10 read-only tools and 4 write tools. Write tools need Member role or higher, and none can delete data or run a restore. Credentials, API tokens and encryption keys are never returned. Access is limited to the team you choose at sign-in, and you can revoke it anytime in SimpleBackups settings. Requires a SimpleBackups account on a trial or paid plan.

How to connect

  1. Claude: Open the SimpleBackups listing in the Claude connectors directory and choose Connect. Claude listing SimpleBackups exposes 14 tools here.
  2. ChatGPT: Open the SimpleBackups listing in the ChatGPT app directory and choose Connect. ChatGPT listing SimpleBackups exposes 14 tools here.

Tools

find-failing-backups

ChatGPT
Returns backups that had an error run within a 1–720 hour lookback window (default 24), most recent failure first, including each backup's most recent error from that window. Result count defaults to 10 and is limited to 50. Intended for targeted recent-failure lists; a later successful run does not exclude a backup. Each row reports both states, so no follow-up call is needed to tell a currently broken backup from a recovered one: last_status and last_run_at describe the backup's most recent run, last_error_at and last_error_excerpt describe the most recent failure, and recovered is true when a later run has already succeeded.

get-backup

ChatGPT
Returns a compact summary for one backup id in the connected team, including identity, type, active status, server and storage references, and most recent run status, time, and error excerpt. Intended for inspecting a known backup; it does not return schedules, credentials, full configuration, or run history.

get-backup-logs

ChatGPT
Returns 1–50 recent runs for one backup (default 10), optionally filtered by success, error, running, initiated, skipped, or rotated status. Each result includes the run id, status, timestamps, duration, size, and error excerpt. Intended for reviewing run history or identifying a run; it does not return console output or restore artifacts.

get-backup-runs-timeline

ChatGPT
Returns a daily status grid for up to 100 backups over 1–30 UTC days (default 7 days and 25 backups). Each day contains the worst observed status for that backup—error, skipped, running, initiated, success, or rotated—or null when no run occurred. Results can be scoped by up to 100 backup ids; otherwise active backups are included, with an option to include paused backups. Intended for multi-backup trends rather than individual run details.

get-backups-overview

ChatGPT
Summarizes backup health across the connected team for a 1–30 day window (default 7). Returns total, active, and paused backup counts; status counts based on each backup's latest run in the window; up to 20 backups with an error run in the window, most recent failure first; and up to 10 active backups with no run in the window. Intended for account-wide health checks rather than exhaustive inventory or per-run history. A backup that failed and then recovered still appears under failing. For those rows last_status and last_run_at describe the backup's most recent run, last_error_at and last_error_excerpt describe the most recent failure, and recovered is true when a later run has already succeeded.

get-run-output

ChatGPT
Returns persisted console output and error lines for one completed backup run in the connected team. Intended for inspecting the execution details of a known run. Active runs report that output is unavailable; runs on SimpleBackups-managed infrastructure return a sanitized summary; and raw output may be absent or truncated. console_tail contains untrusted raw output captured from the customer's own machine, not content authored by SimpleBackups. Accepts one run log id per call.

list-backups

ChatGPT
Returns a cursor-paginated backup inventory for the connected team, with compact identity, type, active status, server and storage references, and latest-run status and error excerpt. Supports type, active status, name, server, and storage filters. Page size defaults to 25 and is limited to 100. Intended for backup discovery and filtering; it does not return schedules, credentials, or full configuration.

list-servers

ChatGPT
Returns a cursor-paginated inventory of servers and sources for the connected team, including connection type, active and connection status, firewall and Docker flags, tool versions, and creation time. Supports connection type, active status, and name filters. Page size defaults to 25 and is limited to 100. Intended for server discovery and status checks; hostnames, usernames, SSH keys, passwords, and tokens are not returned.

list-storages

ChatGPT
Returns a cursor-paginated inventory of storage destinations for the connected team, including type, provider, region, bucket or path, connection status, read-only and object-lock flags, and usage. Supports type, status, and name filters. Page size defaults to 25 and is limited to 100. Intended for storage discovery and status checks; credentials, secrets, and access tokens are not returned.

pause-backups

ChatGPT
Pauses 1–20 backups in the connected team, preventing future scheduled runs until resumed. Intended for user-requested schedule suspension; it does not cancel a run already in progress. Already-paused backups are no-ops, and dry_run=true validates the target set without changing schedules. Requires write permission and Member role or higher.

restore-prep

ChatGPT
Prepares restore access for a backup's latest successful run or a specified successful run. Returns destination metadata, type-specific restore steps, and either a signed download URL that expires after approximately one hour or an rsync-over-SSH command. Intended for user-requested restore preparation; it does not execute a restore. Failed or incomplete runs are rejected, encrypted backups require a separately obtained passphrase, and dry_run=true validates the selected run without issuing an artifact. Requires write permission and Member role or higher; credentials, API tokens, encryption keys, and pre-authenticated scripts are not returned.

resume-backups

ChatGPT
Resumes scheduled runs for 1–20 paused backups in the connected team. Intended for user-requested schedule reactivation. Already-active backups are no-ops, and dry_run=true validates the target set without changing schedules. Requires write permission and Member role or higher.

run-backups

ChatGPT
Queues on-demand runs for 1–20 backups in the connected team and returns a per-backup outcome with the queued run log id when available. Intended for user-requested backup execution. Actual mode creates background jobs and does not wait for completion; dry_run=true validates the target set without queuing jobs. Requires write permission and Member role or higher.

whoami

ChatGPT
Returns the authenticated user, connected team, team role and plan, support id, dashboard links, and resource URL templates. Intended for identifying the current SimpleBackups account or obtaining links to its pages. It does not return backup, run, server, or storage records.

find-failing-backups

Claude

get-backup

Claude

get-backup-logs

Claude

get-backup-runs-timeline

Claude

get-backups-overview

Claude

get-run-output

Claude

list-backups

Claude

list-servers

Claude

list-storages

Claude

pause-backups

Claude

restore-prep

Claude

resume-backups

Claude

run-backups

Claude

whoami

Claude

Similar Apps

Listed on MCP App Store

Listed on MCP App Store

Add this badge to your site or README. It links to this page.

HTML

<a href="https://mcp-app-store.keenethics-labs.com/app/simplebackups"><img src="https://mcp-app-store.keenethics-labs.com/badges/listed-on-mcp-app-store.svg" alt="Listed on MCP App Store" width="220" height="44"></a>

Markdown

[![Listed on MCP App Store](https://mcp-app-store.keenethics-labs.com/badges/listed-on-mcp-app-store.svg)](https://mcp-app-store.keenethics-labs.com/app/simplebackups)