Skip to content
cccskills
FreePublished on GitHub

cloud-sync

Set up or check claude-mem memory syncing with cmem.ai Pro. Follow installer-based sign-in, verify the cloud connection, and review device details, pending writes, the last successful upload, and errors without sharing tokens in chat.

OriginalShow the original description

Set up or check claude-mem cloud sync with cmem.ai Pro. Use when the user says "set up cloud sync", "sync my memories", "cmem pro", "cloud backup", "sync status", or wants their memory database backed up or synced to their cmem.ai account.

See how to install

Good for

  • Set up memory syncing with cmem.ai Pro
  • Check the cloud connection and pending writes
  • Investigate sync errors
  • Refresh the connection through the installer

Files included (1)

  • SKILL.md4.0 KB

SKILL.md (original)

Review the instructions the agent will receive before you install.

Cloud Sync (cmem.ai Pro)

The installed worker syncs through SyncHub. There is one client, one durable operation log, and no separate sync daemon. This skill checks status and points the user to the installer's cmem.ai sign-in, which writes the connection.

Security rule: never ask the user to paste the sync token into this chat, and never put it in a command you write, print it, put it in argv, or log it. A secret pasted into a chat lands in the transcript, which claude-mem itself can capture and sync. The installer writes the token to ~/.claude-mem/settings.json (mode 0600) without it ever passing through this conversation.

1. Check status

Resolve the worker port and query the always-registered status route:

PORT="${CLAUDE_MEM_WORKER_PORT:-$(node -e "const fs=require('fs'),p=require('path'),os=require('os');const uid=(typeof process.getuid==='function'?process.getuid():77);const fallback=String(37700+(uid%100));try{const s=JSON.parse(fs.readFileSync(p.join(os.homedir(),'.claude-mem','settings.json'),'utf-8'));process.stdout.write(String(s.CLAUDE_MEM_WORKER_PORT||fallback));}catch{process.stdout.write(fallback);}" 2>/dev/null)}"
curl -s "http://127.0.0.1:${PORT}/api/sync/status"
  • configured: true and hub.reachable: true → the worker completed an authenticated GET /v1/sync/status against SyncHub. Report deviceId, pending counts, lastFlushAt, lastError, and the Hub head/checkpoint; stop unless the user asked to replace the connection.
  • configured: true and hub.reachable: false → report hub.error and say the SyncHub connection is not verified. A zero pending count or lastError: null is not success because an empty queue performs no push.
  • configured: false → continue.
  • Connection refused, 404, or 503 immediately after restart → retry every three seconds for about 30 seconds before diagnosing the worker.

2. Connect through the installer

Ask the user to run this in their own terminal:

npx claude-mem install

and to choose CMEM Pro when the installer asks. The installer signs them in through the browser (they approve a device code), then writes the sync token, user id and SyncHub URL into ~/.claude-mem/settings.json itself and restarts the worker. Nothing secret is typed into this chat.

If the user cannot run the installer interactively (CI, a remote box), point them to the headless setup guide, https://docs.claude-mem.ai/cmem-pro-headless, and let them apply its manual recipe in their own terminal or editor. Do not collect the values yourself.

The worker mints and persists a device id on first start and defaults the device name to the hostname.

3. Verify

Once the installer finishes, poll the status route every five seconds for up to 30 seconds. If the worker was not restarted, restart it first:

curl -s -X POST "http://127.0.0.1:${PORT}/api/admin/restart"

Success means configured: true, hub.reachable: true, and lastError: null. The local route always makes an authenticated, read-only SyncHub status probe, even when every pending count is zero; it never uses a legacy cmem.ai Pro status route and never appends or advances sync state. Pending counts describe only writes made after the SyncHub launch baseline; setup does not migrate a pre-launch local corpus.

If hub.reachable is false, report hub.error. If lastError is non-null, report it too, and suggest running npx claude-mem install again to refresh the connection. Never include the token.

4. Report

Report device id, pending counts, last successful flush, Hub reachability and checkpoint, and any Hub/flush error. End with this privacy note:

Cloud sync uploads your observation narratives and full prompt text to your cmem.ai account.

Reviews

No reviews yet. Tell others what you thought.

Comments

No login needed. Share how it worked for you, ask a question, or note where it did not run.

No comments yet. Be the first to write one.

Comments are published right away. Do not include personal information such as email addresses or phone numbers. Up to 2 URLs.

0 / 1,000 characters

This form is protected by reCAPTCHA, and the Google Privacy Policy and Terms of Service apply.

More skills from this repository

Turn agent activity and transcript token counts into a cost report organized around completed outcomes. Separate estimates from measured spending, examine waste and rework, and export readable HTML with supporting JSON, CSV, and evidence.

  • Reviewing weekly agent spending and completed work
  • Comparing costs per completed outcome
  • Investigating the cost of loops and rework
thedotmack/claude-mem101.7KUpdated Oct 19, 2026

babysit

FreeSummary

Monitor a pull request through reviews and automated checks, verify reported issues against the code, and make focused fixes. Repeat testing and review checks until actionable comments and unresolved threads are cleared and the request is ready to merge.

  • Monitoring a pull request while automated checks run
  • Verifying and fixing review comments
  • Checking unresolved threads before a merge
thedotmack/claude-mem101.7KUpdated Oct 19, 2026

Delegate a coding task to a subagent that runs build, merges the ready pull request, and continues from the updated main branch into release, connecting implementation and release work in one automated sequence.

  • Delegating a coding task through release
  • Connecting a ready PR merge to release work
  • Running build and release in sequence
thedotmack/claude-mem101.7KUpdated Oct 19, 2026

ccs-align

FreeSummary

Check a local claude-mem worker, refresh viewer-specific caches of important observations, and exclude or restore records without deleting history. Inspect rules across shared, project, and agent layers, then append conflicts to a report.

  • Checking worker health and refreshing important observations
  • Excluding cached records while preserving source history
  • Restoring previously excluded observations to a cache
thedotmack/claude-mem101.7KUpdated Oct 19, 2026

childhood

FreeSummary

Monitor a shipped release by checking published packages, live services, installed plugins, and late CI results or user reports. Verify regressions, prepare authorized fixes, and report release health with evidence and any remaining verification gaps.

  • Checking package contents after publication
  • Verifying changed user flows in production
  • Confirming installed plugins use the shipped version
thedotmack/claude-mem101.7KUpdated Oct 19, 2026

Set up claude-mem in Cursor with a local or remote worker, the process that runs its work. Follow configuration steps for remote inference through cmem.ai or a local observer using the host's login.

  • Installing claude-mem with a local worker in Cursor
  • Connecting Cursor to a remote claude-mem worker
  • Configuring host-login or cmem.ai inference
thedotmack/claude-mem101.7KUpdated Oct 19, 2026

See all skills by thedotmack

Report a problem with this skill