literaf/dsh-research0

dsh-research

AI4Scholar research pack for DeepSeek Harness (dsh): research skills, literature-first workflow guidance and workspace conventions. Pairs with dsh-ai4scholar for search, full text, citations and figures.

包名
dsh-research
版本
0.2.0
许可证
MIT
最近更新
2026年8月16日

安装

$npx -p @deepseek-ai/dsh dsh plugin --profile web add github:literaf/dsh-research

Configuration

The bundle inserts one row (id: research-pack). Override it from your profile's cordis.patch.yml (a patch replaces the whole config, so restate every key you keep):

- id: research-pack
  config:
    skills: true                  # register the bundled skills
    skillNames: []                # register only these; empty means all
    promptGuidance: true          # register the workflow section
    promptOrder: 140
    workspaceConventions: true    # state the workspace layout in the guidance
    papersDir: papers/
    bibliographyFile: references.bib
    notesDir: notes/
    language: zh                  # zh | en — guidance and skill catalog copy

Language

language selects the language of the workflow section and of each skill's catalog entry. The skill bodies stay Chinese in both settings — they are AI4Scholar's originals, kept verbatim so they stay in sync with their published source. Under language: en the guidance says so explicitly and tells the model to carry the procedure out but report back in the language the user is writing in.