zotero paper mdread

    1

    Convert Zotero PDFs to traceable Markdown with deduplicated image attachments using local MinerU parsing.

    Free

    Security scanned

    Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIDeepSeekDeepSeekMuseMuse+22 more

    zotero paper mdread

    Example session with this skill installed

    Process the three PDFs in my "Transformers" collection. Parse them with MinerU, extract images as Zotero attachments, and generate linked Markdown files with the mdread tag.

    • Read your context and instructions
    • Compiled the zotero paper mdread

    I have processed 3 items from the "Transformers" collection. The PDFs were parsed locally using MinerU 4.0.12. Images were extracted, deduplicated via SHA-256, and imported as child attachments. I have also generated and attached the fulltext-linked.md and page_map.md files for each item.

    Zotero Processing Summary

    Item TitlePDF StatusImages ExtractedMarkdown KeyTag Applied
    Attention Is All You NeedSuccess4 (Unique)NK92J2LSmdread
    BERT: Pre-training of Deep...Success2 (Unique)P9WL3M2Amdread
    Language Models are Few-Shot...Success7 (Unique)QW8XN41Bmdread

    Extraction Details

    • Parser: MinerU v4.0.12 (Basic + OCR)
    • Image Handling: All images converted from base64 to stored attachments in Zotero storage.
    • Linking: fulltext-linked.md now uses relative paths (e.g., ../IMAGE_KEY/image.png).
    • Traceability: page_map.md generated with physical page boundaries (1-起算).

    Next steps

    1. Open Zotero and verify the mdread tag on the new Markdown attachments.
    2. Check the page_map.md for coordinate accuracy on the "Attention Is All You Need" figures.
    3. Run the cleanup command if you wish to remove the temporary raw_images from your project directory.

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

    About this skill

    Stop manually extracting text from academic papers. This skill automates the conversion of Zotero PDF attachments into structured, searchable Markdown using local MinerU parsing. It eliminates the mess of base64 images by converting them into linked Zotero attachments, maintaining a clean library while ensuring every claim in your notes can be traced back to a specific page and coordinate.

    What it does

    • Local PDF Parsing converts academic PDFs into Markdown using MinerU with automatic OCR for complex layouts.
    • Smart Attachment Mapping extracts images, deduplicates them via SHA-256, and imports them as native Zotero child attachments.
    • Traceable Indexing generates a page_map.md and page_map.json to link specific text snippets to physical page numbers.
    • Clean Markdown Export creates a fulltext-linked.md that uses relative paths to local Zotero storage instead of bloated embedded data.
    • Batch Processing handles entire Zotero collections or specific selections with resume support for interrupted runs.

    How it works

    1. Scope Discovery: The agent queries your Zotero library for target parent items, PDFs, and their file paths.
    2. Local Extraction: It runs MinerU locally to generate raw Markdown, JSON page maps, and temporary image assets.
    3. Asset Deduplication: Images are hashed to prevent duplicates before being formally imported into the Zotero database.
    4. Relink & Tag: The final Markdown is generated with relative image links, attached to the parent item, and tagged with mdread for easy filtering.

    Frameworks & tools

    Zotero local database, MinerU (v4.0.12+), local OCR models, and standard Zotero attachment storage protocols.

    Why this beats prompting it yourself

    Standard LLMs cannot access your local Zotero storage or run heavy OCR engines like MinerU. This skill bridges the gap between your local file system and your Zotero database, ensuring images are properly stored as database objects rather than broken local file links.

    Use cases

    • Convert a 50-paper Zotero collection into a searchable Markdown knowledge base.
    • Migrate old Markdown notes with embedded base64 images into a clean, linked format.
    • Create a precise page-by-page index for a complex textbook to facilitate RAG workflows.
    • Clean up redundant attachments by deduplicating images across multiple related papers.

    Known limitations

    Does not support cloud-based parsing or remote Zotero API sync; requires local MinerU installation and direct access to the Zotero storage directory.

    Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for Agents & Orchestration.

    Requires local MinerU v4.0.12+ and direct access to Zotero storage. Does not support cloud-only libraries or direct Zotero SQLite modification.

    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

    Free to use

    No account required to browse

    Trust & safety

    Security scan

    Security scan passed 9 days ago

    • Free with an Agensi account

    Permissions

    No special permissions declared or detected

    Listed9 days ago

    What's inside

    Frequently Asked Questions

    This skill automates the extraction of PDF content into high-quality Markdown, specifically solving the problem of bloated base64 image data and broken links by converting visuals into deduplicated Zotero attachments.

    Yes, this skill requires a local installation of MinerU (v4.0.12+) and direct access to your local Zotero storage directory to perform high-fidelity parsing and asset management.

    Absolutely; the 'Traceable Indexing' feature generates page maps that link Markdown snippets to physical page numbers, making it an ideal preprocessing tool for creating accurate RAG datasets.

    The skill uses SHA-256 hashing to identify identical images across your library, ensuring that redundant figures are only stored once as a native Zotero child attachment to save disk space.

    Your purchase includes the agent skill files, configuration templates for local path mapping, and documentation on integrating the workflow with your Zotero database and MinerU environment.

    This skill is designed for local Zotero desktop environments; it does not support Zotero Cloud or remote API syncing because it requires direct file system access for the OCR engine.

    New in Agents & Orchestration

    Free