skills.rest
selfagency
selfagencyCOMMUNITY
Explore all skills by this author →
WebsiteGitHub

qiaomu-opencli-browser

Automates browser navigation, clicking, and typing to extract structured data and fill forms via indexed page states and network-captured JSON APIs.

Updated Mar 11, 2026
Search Tags:#browser automation#authenticated sessions#web interaction#ui extraction#agent tooling#opencli#network api discovery
One-click install
npx skills add https://github.com/selfagency/agentsy --skill qiaomu-opencli-browser-selfagency
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: qiaomu-opencli-browser
Source: https://github.com/selfagency/agentsy/tree/main/.agents/skills/qiaomu-opencli-browser
Command: npx skills add https://github.com/selfagency/agentsy --skill qiaomu-opencli-browser-selfagency

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

It solves the problem of having AI agents reliably interact with websites to retrieve information or complete tasks without manual browsing.

Core Features & Use Cases

  • Structured DOM inspection (no vision screenshots): Use page state with stable element indices to find and operate UI elements efficiently.
  • Reliable interaction primitives: Navigate, click, type, select, scroll, and wait while avoiding brittle or failure-prone techniques like eval-driven clicking.
  • API-first extraction: Use the browser network capture to discover underlying JSON endpoints for higher reliability than DOM scraping.
  • Reusable automation artifacts: Seed and verify site-specific command adapters so recurring tasks become one command.

Use Case: An agent needs to extract the latest items from a dynamic site (or fill a logged-in form) by reusing existing browser sessions, then return structured results without requiring an LLM API key.

Quick Start

Ask the AI agent to open the target website, inspect the page state, and extract the key fields using browser state and safe interaction commands.

Frequently Asked Questions about qiaomu-opencli-browser

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

FAQPage Schema
How do I automate browser tasks for AI agents without using vision screenshots?▼

Automate browser tasks for AI agents by using structured DOM inspection with stable element indices instead of vision screenshots. This approach enables reliable UI element discovery and operation on dynamic SPAs without relying on brittle, screenshot-first logic.

Can I reuse authenticated browser sessions to extract data from dynamic SPAs?▼

Yes, you can reuse authenticated browser sessions to extract data from dynamic SPAs. The tool navigates authenticated pages step-by-step using existing sessions, allowing agents to retrieve information or fill logged-in forms without requiring a separate LLM API key.

What is the best way to discover underlying JSON APIs during web interaction?▼

The best way to discover underlying JSON APIs is by using browser network capture. This API-first extraction method monitors network traces during web interaction, providing higher reliability than traditional DOM scraping by directly fetching structured data endpoints.

How do I safely click and type on dynamic web pages using an automated agent?▼

Safely click and type on dynamic web pages by using state-based element indexing. The agent interacts with UI elements strictly via their assigned indices for actions like click, type, and select, avoiding failure-prone, eval-driven clicking techniques.

How to create reusable automation artifacts for recurring web form completion tasks?▼

Create reusable automation artifacts by seeding and verifying site-specific command adapters. This sedimentation workflow allows recurring web form completion tasks to be recorded, verified, and eventually executed as a single, deterministic command.

Related Skills

Explore More Software Engineering Skills→
aayushsoamaayushsoam
1

dogfood

Automate exploratory QA testing of web applications with browser automation and structured bug reports.

Community
Advanced
pynbj1001pynbj1001
2

Smart Browser

Extract JavaScript-rendered web page agendas and session data to Markdown and JSON.

Community
Intermediate
dabblersportdabblersport

stream-chain

Orchestrate multi-agent sequential workflows by passing output context between prompt-based tasks.

Community
Advanced
kevinpbuckleykevinpbuckley
586

screenshots

Capture Unreal Editor window, viewport, and asset screenshots via ScreenshotService API.

Community
Basic