Socialist-Sister/dsh-collaboration2

dsh-collaboration

Multi-agent collaboration suite for DeepSeek Harness: a user-configured specialist roster with on-demand dispatch (team_call / roundtable), model comparison, and a multimodal vision bridge — models come from the official provider flow, no bundled adapters.

AI 分析

核心用途是实现多 Agent 协同工作。适合需要复杂任务拆解、多模型结果对比或跨领域专家智能体联合办公的高级用户。

パッケージ
dsh-collaboration
バージョン
0.1.0
ライセンス
MIT
最終更新
2026/08/19

インストール

$npx -p @deepseek-ai/dsh dsh plugin --profile web add github:Socialist-Sister/dsh-collaboration

ドキュメント

README 全文を読む ↗

Roster configuration

collaboration-team:
  agents:
    - { id: main, name: 主代理, role: Coordinates and dispatches specialists }
    - { id: planner, name: 规划师, role: Breaks goals into steps, provider: deepseek-official, model: deepseek-v4-flash }
    - { id: reviewer, name: 审查员, role: Reviews code and designs, provider: deepseek-official, model: deepseek-v4-flash }
    - { id: looker, name: 观察员, role: Vision analysis, provider: zhipu, model: glm-4v-flash }
  • provider = a provider ID added in the official Models page; empty = follow the session model (chat-box selector)
  • Give vision identities (e.g. looker) a vision-capable model, or image tasks fail at runtime
  • Changes apply live — no restart needed

Usage examples

ScenarioWhat the main agent does
Parallel auditsteam_call with instances: 2 hires two reviewer clones, one per module
Follow-up questionteam_message to reviewer#1 about session-fixation attacks
Relay an objectionteam_message critic's objection to planner
Specialist asks specialistresearcher calls team_help for looker; you forward the request and relay the answer back
Group deliberationroundtable with planner, reviewer, critic on one topic
Model comparisonmodel_compare deepseek-v4-pro vs zhipu/glm-4.5 on the same prompt
Read an imagevision sends a screenshot to the vision model and returns text analysis