ClawHarbor for Openclaw

A seamless agent-to-agent task API that allows developers to delegate research, analysis, and content creation to specialized human agents without registration.

shadowpigy
v1.0.0
Mar 9, 2026
0
929
0

Install & Download

1. ClawHub CLI

The fastest way to install a skill directly from the registry.

npx clawhub@latest install clawharbor

2. Manual Installation

Copy the skill folder to one of these locations

Global
~/.openclaw/skills/
Workspace
<project>/skills/

Priority: Workspace > Local > Bundled

3. Prompt Installation

Copy this prompt to OpenClaw to install it automatically.

Help me install clawharbor using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).

Prefer to download?

Get the raw skill files in a ZIP archive.

What is ClawHarbor?

ClawHarbor is a specialized skill designed for the Openclaw Skills ecosystem that enables AI agents to offload complex, real-world tasks to a network of human-in-the-loop specialized agents. By bridging the gap between automated coding agents and manual execution, it provides a structured way to handle research, data collection, and content generation through a simple REST API.

The beauty of this skill lies in its friction-less architecture; it requires no API keys or lengthy registration processes. Developers can simply post a natural language description of a task and receive high-quality results, making it an essential tool for those looking to expand the capabilities of their Openclaw Skills implementations.

ClawHarbor Use Cases

  • Delegating deep market research and competitive landscape analysis.
  • Automated data collection and summarization for large datasets.
  • Generating technical documentation and professional content editing.
  • Managing translation and localization tasks within a development workflow.
  • Offloading any task that can be described in natural language but requires human-level nuance.

How ClawHarbor Works

  1. The AI agent reads the API documentation at the ClawHarbor endpoint to understand current capabilities.
  2. A POST request is sent to the tasks endpoint containing a natural language task description.
  3. The system polls the task status until it transitions to a quoted state with a specific price.
  4. The agent presents a payment URL to the user, ensuring the URL fragment is preserved.
  5. Once payment is detected, the agent confirms the transaction via a POST request.
  6. The system polls for the delivered status once the specialized agent completes the work.
  7. The final structured result is retrieved and returned to the primary user or agent workflow.

ClawHarbor Setup

To get started with ClawHarbor as part of your Openclaw Skills toolkit, you can test the endpoint directly using curl:

# Fetch API documentation
curl https://clawharbor.app/v1

# Submit a new task for delegation
curl -X POST https://clawharbor.app/v1/tasks \
  -H "Content-Type: application/json" \
  -d '{"description": "Perform a competitive analysis of top 5 AI coding agents in 2026."}'

ClawHarbor Data Schema & Taxonomy

ClawHarbor uses a lightweight, status-driven data model to manage the lifecycle of a delegated task. Below is the metadata and endpoint structure:

Attribute Description
Task ID Unique identifier for tracking the delegation
Status Lifecycle states: created, quoted, paid, delivered, or rejected
Payment URL Direct link to a Stripe checkout session with essential # fragments
Result Plain text delivery of the completed agent output

All interactions are handled via standard JSON payloads over HTTPS.

ClawHarbor Advanced Features

  • Human-in-the-loop processing for tasks requiring high nuance and technical accuracy.
  • Zero-config integration requiring no persistent authentication or API keys.
  • Native Stripe integration for secure, transparent transaction handling.
  • Status-polling architecture designed for asynchronous multi-agent orchestration.
  • Flexible input handling that accepts any natural language prompt without rigid schema constraints.

SKILL.md


Loading

Related Openclaw Skills

METADATA

Github Stars: 0
forks: 0

Featured*