Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIDeepSeekDeepSeekMuseMuse+22 more

    Page to Markdown

    1

    Convert any web page or JavaScript-heavy site into clean, structured Markdown for local storage and LLM context.

    $2

    Secure checkout via Stripe

    Money-back guarantee

    Converts to your local currency at checkout

    Security scanned
    Page to Markdown

    Page to Markdown

    Example session with this skill installed

    Please fetch the content from https://example.com/blog/ai-engineering-2024 and save it to my research folder as ai-trends.md. I confirm the $0.005 charge.

    • Read your context and instructions
    • Compiled the page to markdown

    I have fetched the page and converted it to clean Markdown. The fetch cost $0.005 USDC and was saved successfully.

    ai-trends.md

    ---
    source: https://example.com/blog/ai-engineering-2024
    retrieved: 2024-10-24
    ---
    
    # The State of AI Engineering in 2024
    
    The shift toward autonomous agents is accelerating. Unlike the prompt-based workflows of 2023, modern engineering focuses on tool-use and structured skill sets.
    
    ## Core Trends
    1. **MCP Integration**: More teams are moving toward standardized Model Context Protocol servers to bridge the gap between models and local tools.
    2. **Small Language Models (SLMs)**: Running specialized models locally for specific tasks like formatting or PII detection.
    3. **Durable Workflows**: Using orchestration layers to manage long-running agentic tasks.
    
    ## Conclusion
    Engineering for AI now requires a deep understanding of infrastructure as much as prompt design.
    

    paid $0.005 · tx 0x742d35Cc6634C0532925a3b844Bc454e4438f44e

    Next steps

    • Open research/ai-trends.md to verify the content formatting.
    • Provide more URLs if you want to continue building your archive.
    ai-trends.md

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

    About this skill

    Scraping modern web pages for an LLM context or personal knowledge base often results in junk: cookie banners, navigation links, and unreadable JavaScript blobs. This skill handles the extraction, cleaning, and conversion of any URL into structured Markdown, delivering only the core content and headings you actually need.

    What it does

    • Cleans web content by stripping ads, scripts, navigation menus, and cookie banners from the source.
    • Converts to Markdown delivering a structured file ready for Obsidian, Notion, or local documentation.
    • Handles JavaScript using a rendering mode for single-page apps (SPAs) and infinite-scroll sites.
    • Manages pay-per-read processing micro-payments via the Base network for high-quality, hosted fetching.
    • Archives metadata automatically adding source URLs and retrieval dates to file frontmatter.

    How it works

    1. Initialize the wallet by configuring the transcribe-mcp server with a private key for a dedicated Base wallet.
    2. Fetch the page by calling fetch_page with the target URL, choosing either a standard fetch or a JS-rendered fetch.
    3. Capture the output as the skill receives the cleaned Markdown content and a cryptographically signed receipt.
    4. Save to disk using the local Write tool to create a .md file with standardized frontmatter.

    Frameworks & tools

    Requires Node 20+ and the transcribe-mcp MCP server (v0.3.0+). Payments are settled in USDC on the Base blockchain using the x402 protocol.

    Why this beats prompting it yourself

    Standard LLM browsing tools often fail on JavaScript-heavy sites or return excessive HTML noise that eats your token window. This skill offloads the fetch to a specialized engine that returns pre-cleaned, high-signal Markdown for a fraction of a cent per page.

    Use cases

    • Research archiving to pull documentation or blog posts directly into a local Obsidian vault.
    • Context injection to provide an agent with clean, readable documentation from a complex web app.
    • Dataset building by converting a list of technical articles into a clean Markdown corpus.

    Known limitations

    Cannot access login-gated or paywalled content. Requests are processed sequentially, not in parallel. Intranet or localhost URLs are unreachable by the remote fetcher.

    Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for LLM Ops.

    Cannot access login-gated or paywalled content. Sequential processing only (no parallel calls). Intranet/localhost URLs are unsupported. Requires USDC on Base.

    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 buying a skill.

    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 27 days ago

    Software
    Transcribe MCP

    Permissions

    Write Files
    Network Access
    Listed27 days ago

    What's inside

    Frequently Asked Questions

    Unlike standard LLM browsing, this skill uses a specialized engine to strip out 'noise' like cookie banners and ads, ensuring you don't waste your token window on irrelevant HTML data.

    The skill includes a dedicated JavaScript-rendering mode specifically designed to handle single-page apps (SPAs) and sites that load content dynamically via infinite scroll.

    You will need Node 20 or higher and the transcribe-mcp server (v0.3.0+); the system uses the Base network for micro-payments to ensure high-quality, hosted page fetching.

    Each fetch automatically generates standardized frontmatter, including the source URL and retrieval date, making it perfect for organizing research in tools like Obsidian or Notion.

    While it can navigate complex layouts and JS-heavy sites, it cannot bypass login screens, paywalls, or access URLs located on a private intranet or localhost.

    New in LLM Ops