DEDUPED Hot AI repos. Real use cases. What people build. Get the Monday 5

How to let your agent read long documents page by page with MinerU

Install the MinerU skill so an agent parses a file, reads it in chunks and keeps stable citation locators.

Short answer: Send your agent the README's install message, which starts with npx skills add opendatalab/MinerU --skill mineru --global --yes; then ask it to read a file; it starts with mineru parse document.pdf --json, which covers the first 10 PDF pages; let it continue with the returned locators instead of re-parsing the whole file. All steps.

  • Research
  • Intermediate
  • 20 minutes
  • From the README
  • Claude Code
  • Codex
  • Cursor

Who it is forA Claude Code or Codex user asking questions about a 300 page contract or book.

The problemDumping a whole long document into context overflows it, and answers come without page references.

The outcomeAn agent that reads progressively, follows continuation locators and cites the page and block it used.

What you need first

  • Python 3.11 to 3.14 in a virtual environment
  • Windows, Linux or macOS; CPU works, compatible GPUs are used automatically
  • Model downloads for the tier you choose
  • Optional: Node.js for npx to install the agent skill

Steps

  1. Send your agent the README's install message, which starts with npx skills add opendatalab/MinerU --skill mineru --global --yes
  2. Ask it to read a file; it starts with mineru parse document.pdf --json, which covers the first 10 PDF pages
  3. Let it continue with the returned locators instead of re-parsing the whole file
  4. Ask it to keep the doc, tier, page and block locators in its answers so every claim is traceable
  5. Expect it to ask before using remote parsing, adding watches or changing persistent config, as the Agent Contract requires

Prompt for your agent

Install the opendatalab/MinerU skill, then read <document> progressively with mineru parse and the returned locators. Answer <question> and cite the page and block locator for every claim.

Caveat. The skill install message also asks the agent to record a global preference for MinerU, so read it before sending if you do not want that.

Use this in your AI agent

Ask your agent

Using the Deduped MCP, get the setup steps for opendatalab/MinerU and "Let your agent read long documents page by page"

Add Deduped to Claude Code once

claude mcp add --transport http deduped https://dedupedai.com/mcp

Cursor, ChatGPT, others

Share: X LinkedIn Hacker News Email Copy link

Third-party repos we don't own. AI-written use cases can be wrong. Disclaimer