zuoyunlai/lunheng-article-pipeline-dsh

论衡(lunheng-article-pipeline)DeepSeek Harness bundle 插件(DSH 适配版)

Bundle verified MIT Unknown v2.1.8-dsh.3
Bundle verified

Listed

0

Skill

Bundle verified

Versionv2.1.8-dsh.3
LanguageUnknown
LicenseMIT
View on GitHub

Preview

Preview 1 of 2: zuoyunlai/lunheng-article-pipeline-dsh
Preview 2 of 2: zuoyunlai/lunheng-article-pipeline-dsh

What it does

Multi-agent deep long-form article pipeline skill for DeepSeek Harness, running 7 roles across 5 phases with parallel literature/data/case retrieval, an evidence base, independent audit and human verification gates.

Best for

  • Teams producing evidence-heavy long-form articles or papers that benefit from parallel literature, data, and case retrieval.
  • Writing workflows that need multiple agent roles for analysis, drafting, counterargument, and independent audit.
  • High-stakes deliverables that can accommodate explicit human verification gates before completion.

Not ideal for

  • Short, routine writing tasks where a five-phase multi-agent pipeline would add disproportionate overhead.
  • Environments without DSH subagent support, which the role orchestration depends on even when role-specific presets are omitted.
  • Low-latency or tightly constrained model-budget workflows that cannot accommodate parallel retrieval, critique, and audit.
  • Fully unattended publishing workflows that cannot provide the documented human verification gates.

README

论衡(lunheng-article-pipeline)— DSH bundle 插件

多 Agent 深度长文流水线技能包,DeepSeek Harness(dsh)bundle 插件版。

把一篇深度文章/论文的生产拆成 8 个角色 + 5 个阶段(Phase 1 为 T1 文献∥T2 数据∥T6 案例 三检索员三方真并行、互不干涉,T6 任何量级必 spawn 含 0 条场景空卡协议;T8 批判伙伴从反方攻击论证;G0-G13 独立审计 + M 门机械化终检),用 dsh subagent 子代理编排,产出有证据底座、反方论证、独立审计、人工核验节点的交付物。

版本:v2.2.8-dsh.1(DSH 适配版,对应正典 v2.2.8)。

安装(在目标机器上)

# 1) 装进 profile 的 node_modules
dsh plugin --profile web add lunheng-article-pipeline

# 2) 把 bundle 加入 profile 清单
#    $DSH_HOME/profiles/web/package.json
#    "dsh": { "profile": { "bundles": ["@deepseek-ai/dsh-base", "@deepseek-ai/dsh-web-app", "lunheng-article-pipeline"] } }

# 3) 重启 dsh web

安装后技能自动出现在会话的 skill 工具目录,无需手动复制到技能根。

内容

  • skills/lunheng-article-pipeline/ — 技能本体(SKILL.md + AGENTS.md + references/:8 张角色卡 + 7 个模板 + _shared/ 共享机制(M 门/F 模式/韧化协议)+ 运行手册 + 设计文档)
  • cordis.patch.yml — bundle 补丁:注册指向包内 skills/ 的 filesystem 技能提供者

按角色分模型(可选)

默认所有角色继承会话模型。若要按角色指派不同模型(检索用便宜快模型,分析/写作/审计用强模型),用随包附带的「分档预设」:

工具 角色 默认模型 环境变量
subagent_retrieval T1 文献 / T2 数据 / T6 案例 deepseek-v4-flash LUNHENG_RETRIEVAL_MODEL
subagent_strong T3 分析 / T4 写作 / T8 批判 deepseek-v4-pro LUNHENG_STRONG_MODEL
subagent_audit T5 审计 deepseek-v4-pro LUNHENG_AUDIT_MODEL
# 1) 复制预设到用户预设根
cp -r examples/preset "$DSH_HOME/.agent-presets/lunheng"

# 2) 新会话在预设选择器里选「论衡分档」

# 3) 换模型:设环境变量后重启 dsh(模型挂载期求值一次)
export LUNHENG_AUDIT_MODEL=claude-opus-5
dsh web

不装预设也没关系——技能会回退到 subagent,所有角色继承会话模型。详见 examples/preset/README.md 与 docs/installation.md。

验证

dsh --profile web --dump-config   # 应能看到 skill-filesystem-lunheng 行
# 新开会话后,skill 工具目录应列出 lunheng-article-pipeline

发布

npm login     # 首次需登录
npm publish   # 或 npm publish --access public

文档

  • docs/installation.md — 安装与验证
  • docs/usage.md — 使用流程(五阶段 + 产物结构)
  • docs/architecture.md — 架构(8 角色 + 三角验证 + G0-G13 审计 + M 门)
  • docs/introduction.md — 插件介绍
  • docs/faq.md — 常见问题
  • CHANGELOG.md — 版本历史
  • CONTRIBUTING.md — 维护与同步指南

许可

MIT,见 LICENSE。

Frequently Asked QuestionsFAQ

Use the verified command dsh plugin --profile default add github:zuoyunlai/lunheng-article-pipeline-dsh in a DSH-enabled shell. The command resolves the public package metadata and keeps the plugin attached to the catalog identity shown on this page.