Manus AI Agent for Openclaw

An autonomous AI agent skill that browses the web, executes complex tools, and delivers complete work products like reports and code.

mvanhorn
v1.3.1
Mar 7, 2026
4
3.2k
0

Install & Download

1. ClawHub CLI

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

npx clawhub@latest install manus

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 manus 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 Manus AI Agent?

Manus is a sophisticated autonomous agent designed to handle end-to-end task execution. By leveraging the Manus API within the Openclaw Skills ecosystem, developers can trigger agents that navigate the live web, utilize specialized tools, and synthesize data into polished final products. Whether you need deep research or complex file generation, this skill provides the bridge between your AI assistant and a fully capable web-browsing operator.

Integrating this into your stack allows for the offloading of time-consuming manual processes. Using Openclaw Skills, you can automate the creation of presentations, technical reports, and software components through a streamlined, API-driven interface that ensures consistency and reliability across different agent profiles.

Manus AI Agent Use Cases

  • Conducting deep market research and generating comprehensive PDF reports.
  • Automating the creation of slide decks and presentations based on web data.
  • Executing multi-step web browsing tasks that require tool interaction.
  • Developing and refining code snippets based on live documentation or web-based repositories.

How Manus AI Agent Works

  1. Initialize a new task by sending a prompt and selecting an agent profile like manus-1.6.
  2. The autonomous agent begins its lifecycle, browsing the web and utilizing tools as required by the prompt.
  3. Monitor the task status periodically through the polling endpoint to track progress from pending to completed.
  4. Extract the resulting data and file URLs from the final output once the agent has finished its execution.
  5. Download the generated assets locally for direct delivery to the end user, ensuring a seamless experience within Openclaw Skills.

Manus AI Agent Setup

To get started, obtain an API key from Manus and configure your environment. Use the following commands to set up the necessary environment variable for Openclaw Skills:

export MANUS_API_KEY='your_api_key_here'

Alternatively, you can define the key in your configuration file under skills.manus.apiKey to authenticate requests to the Manus API.

Manus AI Agent Data Schema & Taxonomy

The Manus skill manages data through task objects and structured output arrays. Below is the organization of a typical task response:

Property Type Description
task_id String The unique identifier for the specific Manus session.
status String Current state: pending, running, completed, or failed.
output Array Contains a list of content objects, including file metadata and URLs.

Files generated by the agent are categorized as output_file types, including a fileName and a direct fileUrl for easy retrieval.

Manus AI Agent Advanced Features

  • Support for multiple agent profiles including Lite for speed and Max for complex reasoning.
  • Adaptive task modes that automatically select the best strategy for the given prompt.
  • Full autonomous agent mode specifically optimized for long-running file creation tasks.
  • Direct extraction and local delivery of output files to bypass unreliable share links.

SKILL.md


Loading

Related Openclaw Skills

Featured*