Setup & Installation
Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:
Version History
Docs: add restart step (/restart) after apply to complete the workflow; clarify restart is gateway-level and kept as separate explicit step.
What This Skill Does
Fetches model lists from OpenAI-compatible provider endpoints and syncs them into the local openclaw.json config. Shows a diff preview before any writes, auto-backs up the config on apply, and prunes stale model aliases from agent defaults by default.
Eliminates manual JSON editing and prevents stale model references from showing up in agent menus after provider updates.
When to use it
- Checking which models a provider added or removed since last sync
- Syncing all configured providers at once after upstream changes
- Adding a new provider to OpenClaw via a guided text wizard
- Keeping agent default model menus aligned with live provider catalogs
- Previewing config changes without touching disk
Example Workflow
Here's how your AI assistant might use this skill in practice.
User sends: /provider_sync provider=cli-usa
- 1Fetch current model list from cli-usa's /v1/models endpoint
- 2Compare fetched models against openclaw.json entries for that provider
- 3Display a diff preview showing added, removed, and changed models
- 4Wait for user confirmation before writing
- 5On confirm with mode=apply, back up config, write updated model list, and prune stale entries from agents.defaults.models
Updated openclaw.json with current provider models; stale aliases removed from agents.defaults.models
Requirements
Accounts, API keys, or tools you or your AI assistant may need to set up while using this skill.