Web to Markdown / JSON, developed by ShiShi, converts web pages into clean Markdown and structured JSON for technical documentation and AI workflows. The extension offers one-click conversion, content cleaning that removes ads and navigation, and a hybrid scraping system that switches between cloud extraction and a local fallback. Export options include copying to clipboard or downloading .md and .json files. The tool targets AI researchers, developers, and technical writers who need LLM-ready content and privacy-oriented processing in a browser workflow.

Who benefits from extracting LLM-ready content?

The extension targets AI researchers, developers, data scientists, technical writers, and students who need clean source material for model training and documentation. It strips away navigation, ads, and footers to expose the main body and outputs either Markdown for readable notes or structured JSON for programmatic ingestion. The design prioritizes machine-ready text, so workflows that feed content into LLMs or RAG systems gain from reduced manual cleanup.

How do Cloud and Local modes affect reliability?

Cloud mode uses the Jina AI Reader API to achieve higher-quality formatting, and it enforces a 20 requests per minute limit that affects batch conversions. Local mode runs entirely in the browser, permitting unlimited attempts and bypassing anti-scraping defenses like Cloudflare, though it may not render dynamic JavaScript as effectively. The extension also avoids backend storage, keeping processing on the client or via encrypted API calls.

How does the extension output and export content for workflows?

Conversion is one click, with a preview and options to copy or download results as Markdown or JSON. The structured JSON includes title, URL, and cleaned content to simplify programmatic ingestion. The extension cannot convert internal browser pages such as chrome:// pages, and it is compatible with Chromium-based desktop browsers. Export choices suit both ad-hoc note capture and automated pipelines.

• Clipboard copy for quick pasting
• .md download for writing and archiving
• .json download for pipelines and tooling

Best suited to focused AI and documentation workflows

The extension fits practitioners who need compact, privacy-minded extraction rather than a full-scale crawler; its niche, generally positive reception among developers and researchers reflects that focus. Expect trade-offs when handling heavily scripted pages or mass conversions, so pair the tool with a dedicated crawler for bulk dataset work. For smaller datasets and archival capture, it serves as an efficient, privacy-conscious utility.