agentbrowseBrowser automation workflows through the agentbrowse CLI for launch, attach, observe, act, extract, navigation, and screenshots.
Install via ClawdBot CLI:
clawdbot install xor777/agentbrowseGrade Fair — based on market validation, documentation quality, package completeness, maintenance status, and authenticity signals.
Calls external URL not in known-safe list
https://github.com/MercuryoAI/agentbrowseAudited Apr 17, 2026 · audit v1.0
Generated Oct 9, 2026
An agent uses AgentBrowse to launch a headless browser, navigate to multiple property listing sites, and extract structured data such as price, address, and square footage. It re-observes pages after pagination or dynamic loading to ensure fresh target refs, then consolidates results into a comparable market report.
A retail analytics agent attaches to a browser session and periodically navigates to competitor product pages to capture pricing and availability. It uses extract with a JSON schema to produce tabular data and screenshots as supporting evidence, flagging anomalies for review.
A software QA agent launches a browser, observes each page to discover interactive elements, and performs click/type/select actions against target refs to simulate real user flows. After each step it re-observes to verify UI changes and stores screenshots for bug reporting and audit trails.
A sales operations agent navigates to trade directories and association member pages, extracts company names, contact details, and categories using schema-guided extraction. The agent stays within public pages and hands off any login-gated content to a protected-flow companion tool.
An agent launches a browser to complete routine online government forms such as permit renewals or license registrations. It observes the form, fills known fields, and pauses when the flow requires identity verification or payment, escalating to the user or a protected-flow tool at that boundary.
A subscription platform that wraps AgentBrowse with scheduling, monitoring, and result dashboards for teams running recurring web tasks. Customers connect their own browser environments or use hosted sessions to run extraction and action workflows at scale.
A service business where analysts use AgentBrowse to collect, clean, and deliver structured web data to clients on a project or ongoing basis. Deliverables include JSON datasets, screenshots, and change-detection reports tailored to each client's schema.
A companion product that integrates AgentBrowse as the browser layer inside a broader AI agent platform, enabling protected login and payment steps through tools like MagicPay. The add-on handles session recovery, observability, and guardrails for enterprise agent deployments.
💬 Integration Tip
Install the CLI globally with npm and run `agentbrowse init <apiKey>` before using AI-assisted observe or extract, since core commands like launch, navigate, act, and screenshot work without a key. Always re-observe after navigation or UI changes instead of reusing stale refs, and treat `close` as teardown only—never as a success signal.
Scored Oct 9, 2026
Meta-skill for AI agent self-improvement. Analyzes runtime logs to detect error patterns, regressions, and inefficiencies, then generates structured improvem...
Stop waiting for prompts. Keep working.
Turn OpenClaw into a learning-loop agent with seeded workspace rules, skill promotion, reflective memory, and proactive maintenance.
Complete toolkit for creating autonomous AI agents and managing Discord channels for OpenClaw. Use when setting up multi-agent systems, creating new agents, or managing Discord channel organization.
Billions decentralized identity for agents. Link agents to human identities using Billions ERC-8004 and Attestation Registries. Verify and generate authentic...
AI agent for finance search agent tasks