backup-setup
Skill509 starsChanged 3 months agoMITView on GitHub
Set up automatic vault backups to a synced folder or a cloud provider, with verified archives and tiered retention. Use when the user says 'back up my vault', 'set up backups', 'where are my backups going', or asks about losing their notes. Not for restoring or testing a restore (`backup-restore`); not for a one-off backup right now (`backup-now`).
- Reads credentials
- Installs packages
1 other project uses a version of this.
Adopted by
Most recently taken up by
- OhJuhun/claude-backup-synctheir own version, first seen Nov 8, 2026
Other versions1 project uses a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 1
- OhJuhun/claude-backup-sync★ 0same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository153 other files it gives its agents.
- AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- .claude/plugins/compound-engineering/CLAUDE.mdCLAUDE.md
- .cursor/rules/search-routing.mdcCursor rule
- anthropic-algorithmic-artSkill
- anthropic-brand-guidelinesSkill
- anthropic-doc-coauthoringSkill
- anthropic-docxSkill
- anthropic-frontend-designSkill
- anthropic-internal-commsSkill
- anthropic-mcp-builderSkill
- anthropic-pptxSkill
- anthropic-slack-gif-creatorSkill
- anthropic-theme-factorySkill
- anthropic-webapp-testingSkill
- anthropic-web-artifacts-builderSkill
- anthropic-xlsxSkill
- atlassian-setupSkill
- backup-nowSkill
- backup-restoreSkill
- calendar-setupSkill
- commitmentsSkill
- create-mcpSkill
- daily-planSkill
- daily-reviewSkill
- decision-logSkill
- delegate-checkSkill
- dex-backlogSkill
- dex-doctorSkill
- dex-improveSkill
- dex-obsidian-setupSkill
- dex-rollbackSkill
- dex-updateSkill
- dex-whats-newSkill
- diff-profileSkill
- enable-semantic-searchSkill
- getting-startedSkill
- goal-backlogSkill
- google-workspace-setupSkill
- granola-setupSkill
- identity-snapshotSkill
- industry-truthsSkill
- initiative-kickoffSkill
- integrate-mcpSkill
- journalSkill
- manage-capabilitiesSkill
- meeting-closeoutSkill
- meeting-prepSkill
- ms-teams-setupSkill
- pipedrive-setupSkill
- pipeline-syncSkill
- process-meetingsSkill
- product-briefSkill
- project-healthSkill
- prompt-improverSkill
- relationship-radarSkill
- reviewSkill
- save-insightSkill
- scrapeSkill
- skill-scoreSkill
- The first 60 shown.
The file
We split this skill into the rules it gives an agent
5 rules. About 0 are common, 3 are distinctive, and 2 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈3
Distinctive · estimate
This author's own choices. Worth a look.
2
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 19
Every run writes a set of three files, stamped with the date and time:
- Distinctive: line 31
Ask which destination suits them, in plain terms:
- Very specific: line 48
If they choose rclone, use a named remote they created with rclone config (for example b2:dex-backups).
- Very specific: line 60
Write to System/integrations/config.yaml under backup::
- Distinctive: line 86
Do not tell them backups are set up until one has actually run.
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
<!-- Generated from `.claude/skills/backup-setup/SKILL.md` by `scripts/generate-agents-skills.py`. Do not edit. -->Backup Setup
Set up (or change) automatic backups of the whole vault: notes, projects, people, tasks, settings, and the vault's own version history. Runs on a schedule in the background, keeps a sensible ladder of older copies, and tells you loudly if it ever stops working.
What a backup contains
Every run writes a set of three files, stamped with the date and time:
| File | What it is |
dex-vault-<stamp>.tar.gz | The whole vault as one compressed archive |
dex-vault-<stamp>.bundle | The vault's version history, individually verified so a restore is provable rather than hoped for |
dex-vault-<stamp>.sha256 | Fingerprints of both, so damage in storage is detectable |
Secrets are deliberately left out: the .env file holding AI keys, generated tool configuration that can reference credentials, saved sign-in tokens (*token.json), the System/credentials folder, any .key or .pem file, virtual environments, and caches. Never store those in a synced folder. Keys live in the system keychain or get re-entered on restore. See docs/backup-restore.md for what a full rebuild needs beyond these files.
Step 1: Where should backups go?
Ask which destination suits them, in plain terms:
- A synced folder (default, simplest): OneDrive, iCloud Drive, Dropbox, Google Drive, or a plugged-in external disk. Anything that looks like a folder on their Mac. The file-sync app carries copies off the machine.
- A cloud storage provider directly: Amazon S3, Backblaze B2, Google Drive proper, and around seventy others, using a tool called rclone. Choose this when they want backups genuinely off their machine rather than mirrored by a sync app, or their vault is large enough that a sync folder is awkward.
For option 1 ask for the folder path. Offer to detect likely candidates:
ls -d ~/Library/CloudStorage/* ~/Dropbox ~/Google\ Drive 2>/dev/nullFor option 2, check whether rclone is present and configured:
rclone listremotesIf they choose rclone, use a named remote they created with rclone config (for example b2:dex-backups). Never put an inline connection string carrying a key or secret into backup.remote: it would be written to the local run record and archived inside the backup itself.
If rclone is missing, say so plainly and give them the choice: install it (brew install rclone, then rclone config to add their provider) or start with a synced folder now and switch later. Do not attempt to configure their cloud credentials for them; rclone's own setup handles that interactively and safely.
Step 2: How many copies to keep?
Explain the default in one line: 7 daily, 4 weekly, 3 monthly. Recent copies for accidents, older copies for problems noticed late (a bad edit that spread quietly, a file corrupted weeks ago). Roughly fourteen sets, which for a typical vault is a couple of gigabytes.
Accept a different ladder if they want one. The newest set is never deleted, whatever the ladder says.
Step 3: Save the settings
Write to System/integrations/config.yaml under backup::
backup:
enabled: true
backend: folder # folder | rclone
destination: /path/to/folder # folder backend
remote: "" # rclone backend, e.g. "b2:dex-backups"
retention:
daily: 7
weekly: 4
monthly: 3Step 4: Schedule it
Install the background job (daily, 12:30 by default; ask if they'd prefer another time, and pick a time their machine is usually awake):
python3 core/backup/install_backup_job.py --hour 12 --minute 30On macOS this schedules the job so missed runs fire when the machine wakes. On other platforms the installer does not pretend: it prints the exact line to schedule the same command with the platform's own scheduler, and installs nothing.
Step 5: Prove it works
Do not tell them backups are set up until one has actually run. Run one now:
python3 core/backup/backup_vault.pyThen show them the real result from System/.dex/backup-last-run.json: the set name, size, and destination. If it failed, show the recorded error and fix the cause before declaring success.
Finish by telling them three things: where backups are going, how to get one on demand (/backup-now), and that /backup-restore in test mode proves a restore works without touching anything.
If backups stop working
Opening Dex warns when the newest successful backup is more than two days old, or the last run failed — the same ⏰ notice used for meeting sync. Ordinary /dex-doctor (no extra scan) reads the last-run record and says what went wrong. If they ask why a warning appeared, read System/.dex/backup-last-run.json and report the recorded error verbatim rather than guessing. A single file that could not be read is named in that record and does not block the rest of the backup.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

