AI Agent Hub
Back to plugins
🧰

dsh-pdf-reader

Web Tools Updated 2026.08.15

Run the following command in DeepSeek Harness:

dsh plugin install ralfsqual/dsh-pdf-reader

Paste the following prompt into your AI chat to install this plugin:

Run dsh plugin install ralfsqual/dsh-pdf-reader in the DeepSeek Harness terminal to install this plugin. The full source code is available at https://github.com/ralfsqual/dsh-pdf-reader . Restart dsh web after installation.

About this plugin

In DeepSeek Harness workflows, agents often need to read PDF documents—contracts, papers, reports—but there is a lack of a secure, lightweight local parsing option.

dsh-pdf-reader registers a read_pdf model tool that extracts text-layer content page by page using pdfjs-dist, entirely on the host. It supports configurable maxPages and maxChars limits, confines file access to the workspace directory, is strictly read-only, and makes zero network requests.

Ideal for developers who need PDF reading in their agent pipelines and prioritize data privacy. Note that it extracts text layers only; scanned or image-only PDFs will yield empty results, as OCR is out of scope.

Use Cases

  • Let an agent read and summarize a PDF report in the workspace during a session
  • Combine with @-file mention plugins so the agent can auto-parse a paper PDF and answer questions
  • In batch documentation pipelines, have the agent extract text layers from PDFs one by one

Best For

  • Developers who need PDF handling in agent workflows and prioritize data privacy
  • Teams that want to avoid uploading documents to third-party services
  • Users building localized document processing pipelines with DeepSeek Harness