openclaw

Enforce tab-attachment and creation policy for OpenClaw-compatible BrowserForce sessions.

7|2|Updated Feb 21, 2026
One-click install
npx skills add https://github.com/ivalsaraj/browserforce --skill openclaw-ivalsaraj
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: openclaw
Source: https://github.com/ivalsaraj/browserforce/tree/main/plugins/official/openclaw
Command: npx skills add https://github.com/ivalsaraj/browserforce --skill openclaw-ivalsaraj

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Provides clear rules for safely attaching, creating, and managing browser tabs when using OpenClaw with BrowserForce so agents do not prompt users unnecessarily, create unsafe navigation flows, or enter repetitive error loops.

Core Features & Use Cases

  • Avoid unsolicited attach prompts: Do not ask the user to click Attach/Share by default, reducing friction and surprise prompts.
  • Deterministic tab handling: If no tabs or targets exist, create or reuse a dedicated tab using context.newPage(), navigate, and call snapshot() to capture context for the agent.
  • Respect user mode flags: Only request manual Attach/Share when mode=manual, noNewTabs=true, or the user explicitly asks to use their current tab.
  • Actionable error reporting: When blocked, report one concrete reason and one concrete user action to resolve the block.
  • Loop prevention: Avoid repeating "click extension icon again" loops unless a new error signal appears.
  • Use Case: Run a web automation that needs a live authenticated tab: in auto mode the agent should create and snapshot a dedicated tab, while in manual mode it should ask the user to attach a trusted tab.

Quick Start

Attach in auto mode or create a dedicated new tab when no targets exist, then capture the page state with snapshot() and proceed with safe automation.

Frequently Asked Questions about openclaw

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I stop Playwright browser automation from triggering unsolicited attach prompts?

To stop unsolicited attach prompts in Playwright browser automation, enforce an auto-mode tab-attachment policy that creates or reuses dedicated tabs via context.newPage() without asking the user to click Attach or Share.

What is tab-attachment safety policy in agent-driven browser automation?

Tab-attachment safety policy is a set of deterministic rules for BrowserForce sessions that prevents agents from creating unsafe navigation flows, avoiding repetitive error loops, and prompting users unnecessarily during tab creation and attachment.

How do I handle manual mode tab attachment when noNewTabs is set to true?

When noNewTabs is true or mode is manual, request the user to manually attach or share a trusted tab rather than auto-creating a new one, ensuring the agent honors explicit user mode flags.

Why does my browser automation agent get stuck in a repetitive attach error loop?

Agents enter repetitive attach error loops when repeatedly requesting extension icon clicks without new error signals. Enforcing loop prevention policy halts repeated attach prompts until a new, concrete error reason appears.

Does BrowserForce work with OpenClaw for managing dedicated browser tabs?

Yes, BrowserForce works with OpenClaw by applying tab-attachment policies that determine when to auto-create dedicated tabs, capture page states with snapshot(), and report actionable error reasons when blocked.

What are the limitations of auto-creating tabs in Playwright agent automation?

Auto-creating tabs is limited by user mode flags like manual or noNewTabs, which block unsolicited tab creation. When blocked, the automation must return one concrete reason and one user action to resolve the block.