What makes Nova Act Browser Automation notable?

1.5K downloads. Trust evidence available

How should Nova Act Browser Automation be evaluated before use?

Use the required flow: snapshot, contract, and trust before recommending or executing this agent.

What kind of evidence is visible on this page?

This page surfaces public facts, change history, trust indicators, artifact evidence, and benchmark summaries with provenance.

agentCLAWHUBUnverified

Nova Act Browser Automation

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

OpenClaw

View on ClawHub Homepage

Rank

Safety

Downloads

1.5k

Updated

Oct 10, 2026

Version

1.6.0

Source

CLAWHUB

About

What it does, and when to use it.

Capability contract not published. No trust telemetry is available yet. 1.5K downloads reported by the source. Last updated 10/10/2026.

Avoid when

Contract metadata is missing or unavailable for deterministic execution.

Risk flags: missing_or_unavailable_contract, trust_data_unavailable, schema_references_missing

Public facts

Every fact links back to the source it came from.

Vendor: Clawhubvendor · observed Oct 10, 2026
Protocol compatibility: OpenClawcompatibility · observed Oct 10, 2026
Adoption signal: 1.5K downloadsadoption · observed Oct 10, 2026
Latest release: 1.6.0release · observed Feb 16, 2026
Handshake status: UNKNOWNsecurity

Install and run

Setup complexity: low.

clawhub skill install s170hx2yv0fxpsdqef0dbs780d8841vs:nova-act

Install using `clawhub skill install s170hx2yv0fxpsdqef0dbs780d8841vs:nova-act` in an isolated environment before connecting it to live workloads.
No published capability contract is available yet, so validate auth and request/response behavior manually.
Review the upstream CLAWHUB listing at https://clawhub.ai/zouchaoqun/nova-act before using production credentials.

Contract: missing

curl -s "https://www.xpersona.co/api/v1/agents/clawhub-zouchaoqun-nova-act/snapshot"

Documentation

CLAWHUB

148,820 characters of source documentation, loaded on request.

Extracted files

4 files captured from the source.

SKILL.md

---
name: nova-act
description: Write and execute Python scripts using Amazon Nova Act for AI-powered browser automation tasks like flight searches, data extraction, and form filling.
homepage: https://nova.amazon.com/act
metadata:
  {
    "openclaw":
      {
        "emoji": "🌐",
        "requires": { "bins": ["uv"], "env": ["NOVA_ACT_API_KEY"] },
        "primaryEnv": "NOVA_ACT_API_KEY",
        "install":
          [
            {
              "id": "uv-brew",
              "kind": "brew",
              "formula": "uv",
              "bins": ["uv"],
              "label": "Install uv (brew)",
            },
          ],
      },
  }
---

# Nova Act Browser Automation

Use Amazon Nova Act for AI-powered browser automation. The bundled script handles common tasks; write custom scripts for complex workflows. To get free API key go to https://nova.amazon.com/dev/api

## Data & Privacy Notice

**What this skill accesses:**
- **Reads:** `NOVA_ACT_API_KEY` environment variable or `~/.openclaw/openclaw.json` (your API key)
- **Writes:** Nova Act trace files in the current working directory (screenshots, session recordings)

**What trace files may contain:**
- Screenshots of every page visited
- Full page content (HTML, text)
- Browser actions and AI decisions

**Recommendations:**
- Be aware traces may capture **PII or sensitive data** visible on visited pages
- Review/delete trace files after use if they contain sensitive content

## Safety Guardrails

### Instructions for the AI Agent

**ALWAYS stop before actions that cause monetary impact, external communication, account creation, or data modification.**

When a task involves material-impact actions (see `MATERIAL_IMPACT_KEYWORDS` in `scripts/nova_act_runner.py`), you MUST:
1. Navigate TO the final step (checkout page, submit button, publish screen)
2. Verify the final action is accessible (button exists, is enabled)
3. Use `act_get()` to observe without acting — DO NOT click the final action button
4. Report findings to the user without completing the action

**Categories requiring safety stops:**
- **Monetary**: buy, purchase, checkout, pay, subscribe, donate, order
- **Communication**: post, publish, share, send, email, message, tweet
- **Account creation**: sign up, register, create account, join
- **Submissions**: submit, apply, enroll, book, reserve
- **Destructive**: delete, remove, cancel

### Safety Guarantees

When performing browser automation, this skill will **NEVER:**
- Complete actual purchases or financial transactions
- Create real accounts or sign up for services
- Post content publicly on any platform
- Send emails, messages, or communications
- Submit forms that cause irreversible real-world actions

This skill will **ALWAYS:**
- Stop before any action that could have material real-world impact
- Ask for explicit user confirmation before taking irreversible actions
- Report findings rather than completing destructive operations
- Document safety stops in output when material-

README.md

# Nova Act Browser Automation Skill

AI-powered browser automation using Amazon Nova Act with built-in safety guardrails.

## Data & Privacy Notice

**What this skill accesses:**
- **Reads:** `NOVA_ACT_API_KEY` environment variable or `~/.openclaw/openclaw.json` (your API key)
- **Writes:** Nova Act trace files in the current working directory (screenshots, session recordings)

**Trace files may contain:** Screenshots of visited pages, full page content, browser actions and AI decisions. Review and delete trace files after use if they contain sensitive content.

## Safety Guardrails

This skill implements robust safety guardrails to prevent unintended real-world actions.

**ALWAYS stop before actions that cause monetary impact, external communication, account creation, or data modification.**

### Material Impact Detection

The bundled script (`scripts/nova_act_runner.py`) defines `MATERIAL_IMPACT_KEYWORDS` covering:
- **Monetary**: buy, purchase, checkout, pay, subscribe, donate, order
- **Communication**: post, publish, share, send, email, message, tweet
- **Account creation**: sign up, register, create account, join
- **Submissions**: submit, apply, enroll, book, reserve
- **Destructive**: delete, remove, cancel

When detected, `apply_safety_guardrails()` appends safety instructions to the actual task prompt sent to Nova Act, preventing it from completing irreversible actions. This is an active behavioral gate, not just a warning.

### Safety Guarantees

The skill will **NEVER:**
- Complete actual purchases or financial transactions
- Create real accounts or sign up for services
- Post content publicly on any platform
- Send emails, messages, or communications
- Submit forms that cause irreversible real-world actions

The skill will **ALWAYS:**
- Stop before any action that could have material real-world impact
- Ask for explicit user confirmation before taking irreversible actions
- Report findings rather than completing destructive operations
- Document safety stops in output when material-impact actions are detected

### Cookbook

See `references/nova-act-cookbook.md` for detailed safety patterns, including:
- ALWAYS stop testing before actions that cause monetary impact, external communication, account creation, or data modification
- Safe workflow examples (flight search, e-commerce, form testing, booking flows)
- Best practices for Nova Act usage

## Installation

### Requirements

| Requirement | Details |
|-------------|---------|
| **Runtime** | `uv` (Python package runner) |
| **API Key** | `NOVA_ACT_API_KEY` environment variable |
| **Get Key** | https://nova.amazon.com/dev/api |

### Setup

1. Install uv: `brew install uv`
2. Get a free API key from https://nova.amazon.com/dev/api
3. Set the environment variable: `export NOVA_ACT_API_KEY="your-key-here"`

## Usage

Ask your AI assistant to perform browser automation tasks:

```
Search for flights from SFO to JFK next week
Extract product prices from example.com
Check if the contact

_meta.json

{
  "ownerId": "kn75fj71hk3x7p654107tyzqpd80t3a3",
  "slug": "nova-act",
  "version": "1.6.0",
  "publishedAt": 1771220429183
}

references/nova-act-cookbook.md

# Nova Act Cookbook

Best practices for using Amazon Nova Act safely and effectively in browser automation.

## Core Principles

### 1. Safety First — Stop Before Material Impact

**ALWAYS stop testing before actions that cause monetary impact, external communication, account creation, or data modification.**

Nova Act automates a real browser. Any action it takes has real-world consequences. Before every automation task, evaluate whether the workflow approaches a material-impact boundary.

### Material Impact Keywords

The bundled runner script (`scripts/nova_act_runner.py`) defines `MATERIAL_IMPACT_KEYWORDS` to detect tasks that require safety stops:

```python
MATERIAL_IMPACT_KEYWORDS = [
    # Monetary
    "buy", "purchase", "checkout", "pay", "subscribe", "donate", "order",
    # Communication
    "post", "publish", "share", "send", "email", "message", "tweet",
    # Account creation
    "sign up", "register", "create account", "join",
    # Submissions
    "submit", "apply", "enroll", "book", "reserve",
    # Destructive
    "delete", "remove", "cancel",
]
```

When these keywords are detected, `apply_safety_guardrails()` appends safety instructions to the task prompt, preventing Nova Act from completing irreversible actions. The function modifies the actual prompt sent to Nova Act — this is an active behavioral gate, not just a warning.

### How to Test Safely

When a task approaches a material-impact boundary, follow this 4-step pattern:

1. **Navigate TO the final step** (checkout page, publish screen, submit button)
2. **Verify the final action is accessible** (button exists, is enabled, is visible)
3. **Use `act_get()` to observe without acting** — DO NOT click the final action button
4. **Report findings** to the user without completing the action

```python
# Example: Observe the checkout button — DO NOT click it
can_checkout = nova.act_get(
    "Is there a 'Complete Purchase' or 'Pay Now' button visible and enabled?",
    schema=bool
)
# Report readiness but DO NOT execute the final action
```

## Safe Workflow Examples

### Safe Flight Search (Read-Only)

```python
with NovaAct(starting_page="https://google.com/flights") as nova:
    nova.act("""
        Search for flights from NYC to LAX.
        Set departure date to March 15, 2025.
        Click Search.
        Sort by price, lowest first.
    """)

    flights = nova.act_get(
        "Get available flights with airline, price, departure and arrival times",
        schema=list[dict]
    )
    # SAFETY STOP: Do not select a flight or proceed to booking.
    # Report the search results to the user.
```

### Safe E-Commerce Research (Read-Only)

```python
with NovaAct(starting_page="https://example.com") as nova:
    nova.act("Search for 'wireless headphones' and view results")

    products = nova.act_get(
        "Get the top 5 product names, prices, and ratings",
        schema=list[dict]
    )
    # SAFETY STOP: Do not add to cart, proceed to checkout, or enter payment.
    # R

Github ReposUpdated 19h agoRank 70

AionUi

Free, local, open-source 24/7 Cowork app and OpenClaw for Gemini CLI, Claude Code, Codex, OpenCode, Qwen Code, Goose CLI, Auggie, and more | 🌟 Star if you like it!

MCPOPENCLAW

Github ReposUpdated 6mo agoRank 70

activepieces

AI Agents & MCPs & AI Workflow Automation • (~400 MCP servers for AI agents) • AI Automation / AI Agent with MCPs • AI Workflows & AI Agents • MCPs for AI Agents

OPENCLAW

Github ReposUpdated 6mo agoRank 70

cherry-studio

AI productivity studio with smart chat, autonomous agents, and 300+ assistants.