Setup & Installation
Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:
Version History
feat: mandatory schema fetch before tool execution; fix JSON output for SDK response objects; add LinkedIn→HarvestAPI provider mapping
What This Skill Does
Centralized OAuth token management for AI agents. Tokens are stored and refreshed by Scalekit, never locally. A single function call returns a fresh, valid bearer token for any configured service.
Instead of implementing per-service token refresh logic and secure storage, agents get a fresh token with one function call, with Scalekit handling the full OAuth lifecycle in the background.
When to use it
- Fetching unread Gmail messages from an agent script without managing token expiry
- Posting to Slack via an AI agent without storing credentials in config files
- Reading GitHub repos from automated scripts using centrally managed OAuth tokens
- Sharing one OAuth connection across multiple agent skills in the same project
- Re-authorizing expired service connections via a prompted link during agent runs
Example Workflow
Here's how your AI assistant might use this skill in practice.
User asks: Fetch my unread Gmail messages
- 1Agent calls get_token("gmail") from scalekit_helper
- 2Scalekit checks if the Gmail connection is configured and active
- 3If not yet authorized, generates a one-minute auth link and prompts the user to click it
- 4User authorizes via browser, Scalekit marks the connection as ACTIVE
- 5Agent receives a fresh access token and calls the Gmail API with it
List of unread message IDs returned from the Gmail API
Requirements
Accounts, API keys, or tools you or your AI assistant may need to set up while using this skill.