Setup & Installation
Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:
Version History
- Version 0.7.2 Optional quality Improvement in API
What This Skill Does
HITL Protocol is an open standard for inserting human decisions into autonomous agent workflows. When a service needs human input, it returns HTTP 202 with a review URL. The agent forwards that URL to the user, polls for the structured result, and continues execution.
It enforces human control over irreversible agent actions using plain HTTP with no SDK, no UI framework, and no custom messenger integration beyond optional native buttons.
When to use it
- Reviewing job listings before an agent submits applications
- Confirming a non-refundable booking before payment goes through
- Approving a generated email batch before it sends
- Selecting a deployment plan before an agent executes it
- Escalating a failed agent action to a human for recovery
Example Workflow
Here's how your AI assistant might use this skill in practice.
User asks: Find senior developer jobs in Berlin and apply to the best ones
- 1Posts search query to job board API
- 2Receives HTTP 202 with a HITL object containing review_url and poll_url
- 3Sends review_url to user: '5 jobs found. Select which to apply for: [url]'
- 4Polls poll_url every 30 seconds until status changes from pending
- 5Reads selected job IDs from result.data and submits applications
Applied to 2 user-selected jobs.