PDF Analysis

Analyze the structure, layout, and content of PDF documents using MinerU. Returns structured output preserving headings, tables, images, formulas, and docume...

This skill appears coherent, but before installing: 1) Confirm the mineru-open-api package source (npm and the GitHub repo) are official and review their README/source so you know whether processing is local or sent to mineru.net. 2) Treat MINERU_TOKEN like any credential: only provide it if you accept that PDFs may be uploaded to the MinerU service; check token scope and storage location used by the CLI. 3) For sensitive documents, prefer local/offline processing or validate that the tool supports fully local extraction; avoid using the 'extract' mode if it uploads data. 4) Install the CLI from the official project pages, inspect what files the installer creates, and run it in a sandbox if possible. 5) Be cautious when running interactive auth in shared environments since tokens are often persisted to disk — locate and audit that config file if needed.

Type: OpenClaw Skill Name: pdf-analysis Version: 0.4.0 The skill bundle provides instructions for using the 'mineru-open-api' CLI tool, an open-source document intelligence engine from OpenDataLab (Shanghai AI Lab), to analyze PDF files. The SKILL.md file contains legitimate installation steps via npm and Go, along with usage examples for document extraction and authentication. No malicious code, data exfiltration attempts, or prompt injection attacks were found in the provided files.