Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    web search page reader

    1

    Live web search plus full-page extraction with no API keys, powered by the companion web-search MCP server.

    $5

    Secure checkout via Stripe

    Money-back guarantee

    Converts to your local currency at checkout

    Security scanned
    web search page reader

    web search page reader

    Example session with this skill installed

    Search for the latest stable version of the React framework and summarize the key changes in the most recent release note.

    • Read your context and instructions
    • Compiled the web search page

    I have searched the web for the latest React release and extracted the content from the official React blog and GitHub repository.

    React v19.0.0 Release Summary

    The latest stable version of React is

    v19.0.0, released on December 5, 2024. This version introduces several major improvements focused on simplifying data handling and improving performance.

    FeatureDescription
    ActionsA new way to handle data mutations and state updates automatically, including pending states and error handling.
    New HooksIntroduction of useActionState, useFormStatus, and useOptimistic to manage form states and UI transitions.
    use APIA new API to read resources like Promises or Context during render, allowing for more flexible data fetching patterns.
    Server ComponentsFull support for Server Components and Server Actions is now integrated into the stable release.
    Document MetadataNative support for rendering <title>, <meta>, and <link> tags anywhere in the component tree.

    Source: React v19 Release Note

    Next steps

    1. Review the breaking changes in the React 19 upgrade guide.
    2. Update your package.json to "react": "^19.0.0".
    3. Replace existing useTransition patterns with the new useActionState hook where applicable.

    Connects securely to your tools. The creator never sees your data.

    About this skill

    Web access for AI agents is often broken by gated APIs, expensive credits, or basic scrapers that get blocked by bot detection. This skill provides a reliable way to search the live web and read full page content using an MCP server that falls back to real browser instances when standard requests fail.

    What it does

    • Live web search queries Bing, Brave, and DuckDuckGo to find current information without requiring individual API keys.
    • Full-page extraction retrieves the readable text from specific URLs, handling complex layouts and dynamic content.
    • Automated browser fallback uses Playwright to bypass basic bot detection and read pages that block standard HTTP clients.
    • Smart result filtering provides quick summaries and snippets first so the agent can select the most relevant sources before committing to a full read.

    How it works

    1. Verification checks for the web-search tools in the agent's environment and initiates a local MCP server installation if they are missing.
    2. Discovery executes get-web-search-summaries to find high-authority URLs and titles related to a specific query.
    3. Extraction calls full-web-search or get-single-web-page-content to pull the actual text content from selected pages.
    4. Synthesis processes the extracted data to answer the user's prompt, citing every external fact with its source URL.

    Frameworks & tools

    The skill requires Node.js 18+ and npm. It uses the Model Context Protocol (MCP) to bridge the agent with a companion server that orchestrates Bing, Brave, DuckDuckGo, and Playwright for browser-based extraction.

    Why this beats prompting it yourself

    Standard agents rely on internal knowledge cutoffs or basic "browse" tools that fail on modern, JS-heavy websites. This skill implements a structured multi-engine search strategy and a robust browser fallback layer, ensuring the agent gets real data even when standard GET requests are blocked.

    Use cases

    • Researching documentation or library changes that occurred after a model's training cutoff.
    • Performing competitor analysis by extracting product features and pricing from live websites.
    • Fact-checking claims against multiple independent primary sources with direct citations.
    • Monitoring current events or release notes to generate up-to-date summaries.

    Known limitations

    Does not bypass paywalls or authenticate into private accounts. Results are limited by the public availability of the target websites and the effectiveness of the underlying search engines.

    Requires manual MCP server installation via Node.js. Cannot bypass paywalls or log in to sites. Extraction may fail on highly restricted or gated content.

    Open docs

    How to use

    1. 1

      Connect your agent to the Agensi MCP

      Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup

    2. 2

      Sign in to Agensi

      Your agent shows a sign-in link. Sign in and click Allow.

    3. 3

      Ask for what you need

      Your agent loads this skill and uses it on your task. It always asks before any payment.

    Reviews

    No reviews yet

    Be one of the first to try it. Every listed skill goes through the trust checks below.

    Security scanned

    Checked by our 8-point scan before listing

    Fresh listing

    Recently published to Agensi

    Money-back guarantee

    Doesn't do what its page promises? You get your money back.

    Trust & safety

    Security scan

    Security scan passed today

    Requires

    playwright
    Node.js
    Web Search MCP
    there.

    Permissions

    Terminal / Shell
    Browser
    Write Files
    Network Access
    Listedtoday

    What's inside

    Frequently Asked Questions

    This skill utilizes a Model Context Protocol (MCP) server that enables your agent to perform live searches and scrape JavaScript-heavy sites using Playwright browser instances, bypassing the need for individual Bing or Brave API keys.

    The skill requires Node.js 18+ and an environment that supports the Model Context Protocol (MCP), such as Claude Desktop or other compatible AI agent frameworks.

    Unlike standard scraping tools that use simple HTTP requests, this skill features an automated browser fallback that uses Playwright to render pages, allowing it to read dynamic content and navigate sites that typically block bot activity.

    Upon purchase, you receive the full MCP server source code, a configuration guide for your specific agent, and the necessary tool definitions to enable search and extraction capabilities immediately.

    The skill includes a multi-engine search strategy that queries Bing, Brave, and DuckDuckGo simultaneously, providing a broader range of results and higher reliability if one engine is temporarily unavailable.

    While the tool is highly effective at bypassing bot detection on public sites, it cannot bypass subscription paywalls or log into private accounts that require user authentication.

    New in Research & Analysis