Formula Ocr

OCR and recognize mathematical formulas from PDFs and images using MinerU. Converts printed or handwritten equations into structured LaTeX or text representa...

Security Scan

The skill's requirements and instructions are consistent with its stated purpose (formula OCR via the MinerU CLI); nothing requested appears disproportionate or unrelated.

Purpose & Capability

Name/description request OCR of mathematical formulas. The skill only requires the mineru-open-api CLI and a MINERU_TOKEN, which are directly relevant to invoking MinerU's service—no unrelated credentials or binaries are requested.

Instruction Scope

SKILL.md instructs the agent to run mineru-open-api commands (extract, auth) on local files or URLs and to provide MINERU_TOKEN. It does not ask the agent to read unrelated files, other environment variables, or exfiltrate data to unexpected endpoints; the only external endpoint referenced is MinerU (mineru.net).

Install Mechanism

Install paths are standard: npm package (mineru-open-api) or go install from the project's GitHub repo. This is appropriate for a CLI-only integration. Note: npm packages (and go installs from remote repos) can run code at install time, so you should review the package/repo if you require a higher trust level.

Credentials

Only MINERU_TOKEN is required and declared as the primary credential. That single API token is proportional to a service that requires authentication. No other secrets or unrelated environment variables are requested.

Persistence & Privilege

The skill does not request always:true or any elevated persistent presence, and it does not indicate modifying system or other-skill configuration. Autonomous invocation (default) is allowed but not exceptional here.

Assessment

This skill appears coherent, but check a few practical points before installing: 1) Verify the mineru-open-api npm package or the GitHub repo are the official MinerU distribution and review their source if you have security concerns (npm installs can run install-time scripts). 2) Be aware that using the skill will send PDFs/images to MinerU's service (mineru.net) — do not upload sensitive or confidential documents unless you trust the service and its privacy/retention policies. 3) MINERU_TOKEN grants access to your account; treat it like any API secret (store it securely and revoke/regenerate it if compromised). 4) If you require higher assurance, run the CLI in an isolated environment or review the repository code before installing. Overall the requests and instructions match the stated OCR purpose.

Convert and extract content from .pdf / images (.png/.jpg/.jpeg/.webp) using MinerU (mineru-open-api).

Install

npm install -g mineru-open-api
# or via Go (macOS/Linux):
go install github.com/opendatalab/MinerU-Ecosystem/cli/mineru-open-api@latest

Quick Start

# Extract formulas from PDF (requires token)
mineru-open-api extract paper.pdf -o ./out/

# With VLM for better formula accuracy
mineru-open-api extract paper.pdf --model vlm -o ./out/

Authentication

Token required for extract and crawl:

mineru-open-api auth            # Interactive token setup
export MINERU_TOKEN="your-token" # Or via environment variable

Create token at: https://mineru.net/apiManage/token

Capabilities

• Supports local files and URLs
• Requires token (mineru-open-api auth or MINERU_TOKEN env)
• Supported input: .pdf / images (.png/.jpg/.jpeg/.webp)
• Language hint with --language (default: ch, use en for English)
• Page range with --pages (where applicable)

Notes

• Formula recognition requires extract with token. The --formula flag is enabled by default.
• Output goes to stdout by default; use -o <dir> to save to file
• Binary formats (docx) require -o flag (cannot stream to stdout)
• All progress/status messages go to stderr
• MinerU is an open-source project by OpenDataLab (Shanghai AI Lab): https://github.com/opendatalab/MinerU