AI Agent Hub
Back to plugins
⚙️

dsh-paper-survey

Workflow Updated 2026.08.25

Run the following command in DeepSeek Harness:

dsh plugin install zeros335882878/dsh-paper-survey

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

Run dsh plugin install zeros335882878/dsh-paper-survey in DeepSeek Harness to install this plugin; the full source is available at https://github.com/zeros335882878/dsh-paper-survey

About this plugin

Turning several papers into a coherent literature-survey deck is one of the most repetitive and time-consuming parts of research writing: close-reading each paper, extracting contributions, aligning them to research questions, and assembling polished slides. dsh-paper-survey codifies this into a three-step workflow—structured interpretation, user confirmation, and A/B-layout deck generation—so the agent handles the full pipeline from PDF to finished presentation.

The plugin ships three collaborating skills: paper-interpret enforces a five-block reading framework where every design choice is justified against a research question; paper-survey orchestrates the main flow and enforces the rule that no slide is produced until interpretation is approved; paper-deck specifies the A/B two-page layout, verbatim-carry-over principle, and pptxgenjs typesetting conventions, guaranteeing camera-ready output.

It is built for graduate students, research groups, and anyone who must present multi-paper comparisons to advisors or reviewers. A single natural-language prompt in a DSH session triggers the entire pipeline, delivering a structurally consistent, properly formatted survey deck so researchers can focus on the analysis and writing that actually matter.

Use Cases

  • Compiling multiple papers into a structurally consistent survey deck
  • Interpreting each paper with a fixed five-block framework aligned to research questions
  • Generating A/B two-page layout presentations via pptxgenjs

Best For

  • Graduate students and postdocs writing literature surveys
  • Researchers presenting multi-paper comparisons to advisors or reviewers
  • Research groups automating repetitive deck-production workflows