dsh-deepread
Run the following command in DeepSeek Harness:
dsh plugin install xiehuan123/dsh-deepread
Paste the following prompt into your AI chat to install this plugin:
Install the plugin in DeepSeek Harness by running dsh plugin install https://github.com/xiehuan123/dsh-deepread.
About this plugin
In an age of information overload, deeply reading long articles, PDFs, and entire books is time-consuming and often misses critical arguments. DeepRead takes an evidence-first approach, turning any long text into traceable claims, evidence, confidence levels, and knowledge maps—avoiding hallucinated summaries typical of AI tools.
Its core capabilities include five reading modes: quick, deep, knowledge map, Feynman technique, and whole-book reading, each tailored to different scenarios. The knowledge-map mode extracts core questions and conclusions, ten content categories, claim-evidence pairs, key data tables, eight relation labels, and four confidence levels; the Feynman mode uses an 11-step loop with spaced repetition to help you truly master the material.
It supports WeChat article URLs, local files (including built-in PDF extraction), and pasted text, with export to Markdown, XMind-compatible mindmaps, or HTML reports. Batch comparison, cost preview, progress transparency, and self-calibrating speed estimates make long-document processing more predictable.
DeepRead is ideal for researchers, students, content creators, and anyone who needs fact-checking before writing or making decisions. Whether you want a quick overview or rigorous evidence verification, it delivers traceable, exportable reading results.
Screenshots
Use Cases
- Read long WeChat articles and quickly extract core arguments and evidence
- Perform structured deep reading on PDFs or entire books and generate knowledge maps
- Fact-check articles and annotate confidence levels before citing them
Best For
- Researchers and students who need to process long texts efficiently
- Content creators who value evidence traceability and want to avoid AI hallucinations
- Analysts who need to compare multiple documents and export reports
Related Plugins
ModLens is a vision plugin for DeepSeek Harness that gives text-only models sight by reading images pasted directly into chat, with zero-config setup and multiple vision engines.
On-demand vision for text-only DeepSeek Harness agents: built-in free keyless vision chain and 14 vision tools, routing image turns as tool calls to vision models with pixel fidelity, no Python needed, one-command install.
Give text-only models in DeepSeek Harness eyes, enabling image Q&A, long-screenshot OCR, UI restoration, and GUI visual tasks.
An armor-breaking plugin for DeepSeek: rewrites instruction priority, expands refusal-blocked phrases, persists across sessions, and shows a green active indicator.