Setup & Installation
Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:
Version History
Initial release. - Introduced Intercom Conversations (Read) skill for Clawhub. - Supports reading Intercom conversations using actions: list, find, and search. - Inputs and outputs follow a clear JSON contract. - Added env var requirement: INTERCOM_ACCESS_TOKEN. - Source metadata provided in clawhub.skill.json; removed package-lock.json.
What This Skill Does
Reads Intercom conversations via three actions: list, find by ID, and search with filters. Returns conversation data including pagination cursors for iterating large result sets.
Querying Intercom's API directly avoids exporting CSVs or navigating the UI when you need conversation data programmatically.
When to use it
- Pull all open conversations assigned to a support agent
- Fetch a specific conversation by ID to summarize it
- Search conversations by customer email or tag
- Page through hundreds of conversations for a weekly report
- Find conversations mentioning a specific product or error message
Example Workflow
Here's how your AI assistant might use this skill in practice.
User asks: Find all conversations from the last week mentioning 'billing error'
- 1Call conversations.search with a query filtering on keyword 'billing error'
- 2Receive paginated results with conversation summaries
- 3If next_starting_after cursor is present, call search again to fetch next page
- 4Collect all matching conversations across pages
- 5Return list of conversation IDs and excerpts to user
A list of conversation IDs and relevant snippets matching 'billing error'
Requirements
Accounts, API keys, or tools you or your AI assistant may need to set up while using this skill.