ClawKit Logo
ClawKitReliability Toolkit
Back to Registry
Official Verified browser automation Safety 3/5

nova-act

Write and execute Python scripts using Amazon Nova Act for AI-powered browser automation tasks like flight searches, data extraction, and form filling.

Why use this skill?

Automate complex browser tasks, flight searches, and data extraction with Amazon Nova Act. Integrate AI-powered web navigation and structured data scraping into your OpenClaw agent.

skill-install — Terminal

Install via CLI (Recommended)

clawhub install openclaw/skills/skills/zouchaoqun/nova-act
Or

What This Skill Does

Nova Act is a powerful browser automation skill that integrates the Amazon Nova Act engine into your OpenClaw agent workflow. It enables your AI to navigate the web, interact with complex UI elements like dropdowns and search inputs, and perform sophisticated data extraction using natural language. Unlike standard scraping tools, Nova Act "sees" the browser, allowing it to handle dynamic content, login screens, and multi-step form submissions with high reliability. You can use it via the bundled runner for quick, one-off tasks, or build custom Python automation scripts using its robust API to handle complex business logic.

Installation

To install this skill, use the clawhub CLI tool. Run the following command in your terminal:

clawhub install openclaw/skills/skills/zouchaoqun/nova-act

Ensure you have uv installed, as it is required to manage the Python environment and dependencies automatically. You will also need an API key from https://nova.amazon.com/dev/api to authorize the agent's browser operations.

Use Cases

Nova Act is best suited for:

  • Automated Research: Pulling pricing, availability, or market data from airline and hotel booking sites.
  • Data Extraction: Scraping unstructured web content into structured Pydantic models for database ingestion.
  • Administrative Workflows: Filling out web forms, handling multi-page registration flows, or navigating through SaaS dashboards.
  • Quality Assurance: Automating repetitive user-flow tests to ensure web elements function as expected across different states.

Example Prompts

  1. "Open Amazon Flights, find me the cheapest round-trip flight from SFO to JFK for next week, and save the result to a JSON list."
  2. "Go to the LinkedIn jobs page, search for 'Senior AI Engineer' in San Francisco, and extract the top 10 company names and job titles."
  3. "Fill out the contact form on example-site.com using these details: Name: Jane Doe, Email: [email protected], and ask about current pricing."

Tips & Limitations

  • Combine Steps: Always combine sequential browser actions into a single nova.act() call. This allows the underlying model to maintain a unified context of the page state, significantly reducing errors.
  • Use Schema Extraction: When extracting data, always define a Pydantic model. This ensures that the data you receive is clean, typed, and ready for further processing.
  • Authentication: For sites requiring MFA or private credentials, consider using the custom script flow to manage session cookies securely rather than trying to handle interactive login flows every time.
  • Limitations: Nova Act relies on screen-state analysis; pages with extremely high-frequency dynamic updates or non-standard canvas-based UI may occasionally cause the agent to lose its place. Ensure your task logic is granular when dealing with non-standard web architectures.

Metadata

Stars879
Views1
Updated2026-02-11
View Author Profile
AI Skill Finder

Not sure this is the right skill?

Describe what you want to build — we'll match you to the best skill from 16,000+ options.

Find the right skill
Add to Configuration

Paste this into your clawhub.json to enable this plugin.

{
  "plugins": {
    "official-zouchaoqun-nova-act": {
      "enabled": true,
      "auto_update": true
    }
  }
}

Tags(AI)

#browser-automation#web-scraping#python#ai-agent#workflow-automation
Safety Score: 3/5

Flags: network-access, file-write, file-read, external-api, code-execution