What does this skill do, and when should you use it?
This skill sets up or checks the cloud sync feature of claude-mem. It guides the user to obtain a sync token, user ID, and SyncHub URL from cmem.ai, then writes them into the local settings file. The skill verifies the connection status and provides diagnostics. The core purpose is to back up memory data securely.
The skill performs these operations: 1. Queries the local worker's /api/sync/status endpoint to check sync configuration and hub reachability. 2. Requests three connection values from the user (sync token, user ID, SyncHub URL). 3. Writes these values into ~/.claude-mem/settings.json using a Node.js script, ensuring file mode is 0600. 4. Restarts the worker via /api/admin/restart and polls status until success. 5. Reports device ID, pending counts, last sync time, hub reachability, and any errors.
- When the user says 'set up cloud sync' or 'sync my memories' to initialize the sync.
- When the user says 'sync status' or 'cloud backup status' to check sync health.
- When the user needs to back up their memory database to their cmem.ai account.
- When sync fails and the user needs to verify their connection values.
How do you install this skill?
- This skill uploads observation narratives and full prompt text to cmem.ai account; users should be fully aware of the privacy implications of data upload.
- Cloud sync depends on cmem.ai Pro paid service; functionality may be affected by network environment, and mainland China users may face accessibility issues requiring additional network configuration.
- Settings file contains sensitive information like sync token; although the skill requires saving with 0600 permissions, users should ensure local environment security.
- Static review cannot verify actual script execution; carefully check placeholder replacement and path assumptions before actual deployment.
- Shell / CLI
- Network access
- Local filesystem
Node.js (for node -e and node - <<)curl
This skill is part of the claude-mem plugin. Install the whole collection: npx claude-mem install or /plugin marketplace add thedotmack/claude-mem then /plugin install claude-mem. The skill is located at plugin/skills/cloud-sync/SKILL.md.
tmp="$(mktemp -d)"
git clone --depth 1 https://github.com/thedotmack/claude-mem.git "$tmp"
mkdir -p ~/.claude/skills
cp -R "$tmp/plugin/skills/cloud-sync" ~/.claude/skills/
rm -rf "$tmp"Generated from the source repository and skill path; it copies only this skill's folder. If the author's install steps above differ, follow those first. To scope it to one project, replace ~/.claude/skills with that project's .claude/skills.
How do you use this skill?
Once installed, send your agent any of these to trigger it:
- set up cloud sync
In a Claude Code session, simply say a trigger phrase like 'set up cloud sync' or 'sync status'. The skill will guide you through obtaining the connection values and writing them to settings.
What are this skill's strengths and limitations?
- Automates the configuration process, reducing manual editing.
- Handles tokens securely, never printing or logging them.
- Provides detailed diagnostics to troubleshoot problems.
- Only works with cmem.ai Pro accounts, not free tier.
- Requires manually copying three values from the website, somewhat tedious.
- No test coverage mentioned; potential unverified edge cases.
How does this skill compare with similar options?
Side by side with related skills; every score comes from the same FSRS standard.
| Skill | FS score | Stars | Last updated | License |
|---|---|---|---|---|
| claude-mem Cloud Sync this page | 42 · Not recommended | ★ 98k | 3d ago | Apache-2.0 |
| Varlock Secrets Guard Skill | 51 · Use with care | ★ 33 | 7mo ago | MIT |
| GKE Production Readiness Review ✓ Google · Official | 49 · Use with care | ★ 21k | 3d ago | Apache-2.0 |
| Hookify Rule Author ✓ Anthropic · Official | 45 · Not recommended | ★ 38k | 3d ago | Apache-2.0 |
| cmux Cloud VM Skill | 56 · Use with care | ★ 28k | 1d ago | NOASSERTION |
How did FollowSkills review this skill?
The skill clearly states security rules: never print or log the sync token, save it with 0600 permissions in settings, and preserve unrelated settings. It uses local HTTP endpoints and environment variables, requesting no global file permissions. However, static reading cannot verify actual execution, and as a paid cloud sync feature, user data (observation narratives and prompt text) is explicitly uploaded to cmem.ai account, though documented, external dependency implies privacy implications for user. Deduction: data flow is transparent but depends on external commercial service, and there is no rollback or revoke mechanism; least privilege is decent but the external data flow is core functionality requiring explicit user consent.
Steps are detailed, with clear status checks, error handling, and verification paths. But static review cannot verify correctness of scripts, and no tests targeting this skill path are seen. Deduction: lack of executed tests and edge-case coverage; error messages are present, but reliability evidence is thin.
Use case is clear (when user explicitly asks to set up or check cloud sync), and trigger semantics are precise. However, cloud sync depends on cmem.ai Pro account and overseas services that may be unreachable from mainland China; adaptability is limited. Deduction: function depends on external commercial service and may be network-restricted; no declaration of adaptation for mainland users; boundary conditions partially described, but environment fit evidence is limited.
SKILL.md is well-structured with sections, security rules, clear scripts and output format. However, versioning, changelog, FAQ, and detailed known-limitations disclosure are missing. Deduction: lack of version history, troubleshooting, and limitation disclosure; maintenance responsibility visible via README, but update path for the skill itself is not clear.
Skill goal is clear and can accomplish core tasks of setup or status check with structured report. But effectiveness evidence is insufficient; no actual output examples or test verification. Deduction: static review cannot verify actual execution effectiveness; output format clear, but evidence of direct usability limited.
Repository has CI and tests, but not covering this skill path. Claims of operation are not independently verifiable. Deduction: lack of traceable evidence specific to this skill, such as CI results or test coverage; third-party execution evidence insufficient.
Open a dimension to read why it scored that way
Evidence confidence：Low — Mostly static review, author material or a limited demo; useful for discovery, not high-risk decisions.
See the full review method →