Duplicacy

Duplicacy Web Edition Skill

Manage Duplicacy Web Edition backups, schedules, storage, and restores for the PHATT-RAID Unraid server.

Trigger Phrases

• Duplicacy
• backup status
• backup schedule
• prune (backup context)
• check backups
• restore from backup
• Wasabi backup

Base URL & Auth

• URL: http://10.0.0.100:3875
• Auth: NONE — no API key, no tokens, no headers
• Protocol: All endpoints POST JSON, except /show_log (GET)
• Env override: DUPLICACY_URL=http://... to point at a different host

Invocation

cd /root/.openclaw/workspace/skills/duplicacy
npx tsx scripts/duplicacy.ts [command] [subcommand] [options]

No args → prints full JSON command tree.

Command Reference

backup

npx tsx scripts/duplicacy.ts backup status
npx tsx scripts/duplicacy.ts backup start 0
npx tsx scripts/duplicacy.ts backup stop 0
npx tsx scripts/duplicacy.ts backup revisions 0
npx tsx scripts/duplicacy.ts backup files 0 --revision 42 --path /

schedule

npx tsx scripts/duplicacy.ts schedule list
npx tsx scripts/duplicacy.ts schedule show 0
npx tsx scripts/duplicacy.ts schedule pause 1
npx tsx scripts/duplicacy.ts schedule resume 1
npx tsx scripts/duplicacy.ts schedule create --name my-schedule --start-time 02:00 --frequency weekly --days "Tue,Thu,Sun"
npx tsx scripts/duplicacy.ts schedule add-job 0 --repo 0 --type backup
npx tsx scripts/duplicacy.ts schedule remove-job 0 2

storage

npx tsx scripts/duplicacy.ts storage list
npx tsx scripts/duplicacy.ts storage info 0
npx tsx scripts/duplicacy.ts storage delete 0

restore

npx tsx scripts/duplicacy.ts restore status
npx tsx scripts/duplicacy.ts restore start --repo 0 --revision 42 --path /
npx tsx scripts/duplicacy.ts restore stop

logs

npx tsx scripts/duplicacy.ts logs recent
npx tsx scripts/duplicacy.ts logs show backup-20260315-020001.log
npx tsx scripts/duplicacy.ts logs show prune-20260315-060001.log --tail 100

dashboard

npx tsx scripts/duplicacy.ts dashboard

settings

npx tsx scripts/duplicacy.ts settings show
npx tsx scripts/duplicacy.ts settings update --key report_url --value https://hooks.example.com/webhook
npx tsx scripts/duplicacy.ts settings test-email

Current Setup (PHATT-RAID)

Known Issues

Wasabi Prune — 403 Forbidden

The all-backups-prune schedule fails with 403 Forbidden when Duplicacy tries to fossilize chunks (a prune step that marks old chunks for deletion before actually removing them).

Root cause: Wasabi's Object Lock or lifecycle policies conflict with Duplicacy's fossil management, which requires certain S3 operations (like creating/deleting .fsl files or running PutObject on specific prefixes) that Wasabi blocks.

To investigate:

# Get the most recent prune log
npx tsx scripts/duplicacy.ts logs recent
npx tsx scripts/duplicacy.ts logs show prune-YYYYMMDD-HHMMSS.log

Workaround options:

1. Disable Wasabi's versioning/Object Lock on the bucket
2. Add the --exclusive flag to prune job options (uses lock-based instead of fossil-based pruning)
3. Create a Wasabi IAM policy granting full S3 permissions on the bucket
4. Use --delete-incomplete in prune options

Multi-Step Workflows

Check Backup Health

# 1. Full overview
npx tsx scripts/duplicacy.ts dashboard

# 2. If issues found, check specific job status
npx tsx scripts/duplicacy.ts backup status

# 3. Get recent logs for any failed jobs
npx tsx scripts/duplicacy.ts logs recent
npx tsx scripts/duplicacy.ts logs show <logname>

# 4. Check schedule health
npx tsx scripts/duplicacy.ts schedule list

Investigate Failed Prune

# 1. Find the most recent prune log
npx tsx scripts/duplicacy.ts logs recent

# 2. View the prune log (filter for errors)
npx tsx scripts/duplicacy.ts logs show prune-20260315-060001.log

# 3. Check if prune schedule is paused (it may auto-pause on repeated failures)
npx tsx scripts/duplicacy.ts schedule list

# 4. Resume the schedule if needed after fixing root cause
npx tsx scripts/duplicacy.ts schedule resume 2

Add New Backup Job

# 1. List existing schedules and repos
npx tsx scripts/duplicacy.ts schedule list
npx tsx scripts/duplicacy.ts backup status

# 2. Create a new schedule (or add to existing)
npx tsx scripts/duplicacy.ts schedule create \
  --name "new-backup-schedule" \
  --start-time 03:00 \
  --frequency weekly \
  --days "Mon,Wed,Fri"

# 3. Add backup job to the schedule (get schedule index from step 2 output)
npx tsx scripts/duplicacy.ts schedule add-job <schedule-index> \
  --repo <repo-index> \
  --type backup

# 4. Verify
npx tsx scripts/duplicacy.ts schedule show <schedule-index>

Restore a File

# 1. Find the right revision
npx tsx scripts/duplicacy.ts backup revisions 0

# 2. Browse files in that revision
npx tsx scripts/duplicacy.ts backup files 0 --revision 42 --path /

# 3. Start restore
npx tsx scripts/duplicacy.ts restore start \
  --repo 0 \
  --revision 42 \
  --path /path/to/file

# 4. Monitor
npx tsx scripts/duplicacy.ts restore status

Agent Usage (Piped/Non-TTY)

All commands output structured JSON when piped:

npx tsx scripts/duplicacy.ts backup status | jq '.result.backup_status'
npx tsx scripts/duplicacy.ts schedule list --json

Response envelope:

{
  "ok": true,
  "command": "backup status",
  "result": { ... },
  "next_actions": [
    "backup start <repo-index>  — trigger a backup manually",
    "logs show <logname>  — view detailed log"
  ]
}

Error envelope:

{
  "ok": false,
  "command": "backup status",
  "error": "connection refused",
  "fix": "Check that Duplicacy Web is running at http://10.0.0.100:3875",
  "next_actions": ["Check server: curl -s http://10.0.0.100:3875/"]
}