DeepSeek Harness 插件

DSH-Subagent-Model-Router

把委派给子代理的任务路由到配置好的模型:subagent_model 工具的 model 枚举即配置的别名,每条路由的标签与使用说明会注入工具 schema 与系统提示;wait-for-subagents 汇合工具等待未完成的可续跑后台子代理并返回其最终结果;在子代理头部与目录行显示当前模型标签;另有 Web 设置页、命名空间内的配置工具与引导设置技能用于管理路由。

跳到安装方式

来源信息

GitHub 仓库
CypherNaught-0x/DSH-Subagent-Model-Router
最近更新
2026年8月19日
分类
模型与服务商
GitHub stars
0

安装

默认先复制一段 Prompt,让 Agent 读页面和仓库;需要自己装时再切到命令。

复制这段 Prompt,发给 DSH、Codex 或其他 Agent,让它先读页面和仓库。

请先不要安装。阅读这个 DeepSeek Harness 插件,说明它解决什么问题、会访问哪些文件、网络或密钥,以及如何安装和卸载。

插件页面:https://deepseekplugins.org/zh/plugins/CypherNaught-0x/DSH-Subagent-Model-Router
GitHub:https://github.com/CypherNaught-0x/DSH-Subagent-Model-Router
插件名:DSH-Subagent-Model-Router
作者:CypherNaught-0x
安装命令:dsh plugin --profile web add github:CypherNaught-0x/DSH-Subagent-Model-Router

确认前不要执行安装命令。

检查来源文件

安装前先看这个插件目录里的 README 和其他文件。

文件资源管理器3 个文件
README.md来源说明 · 只读预览

dsh-subagent-model-router

A DeepSeek Harness Cordis plugin for delegating subagent work to a configured model route that can differ from the parent agent's current model.

What it adds

  • subagent_model: a delegation tool whose model argument is restricted to user-configured aliases.
  • wait-for-subagents: a join tool that waits for this agent's outstanding continuable background children, including standard and model-routed delegations, and returns their terminal results.
  • Per-model tags and routing descriptions embedded in the tool schema and system prompt.
  • model_subagent_catalog: a read-only view of models advertised by registered LLM providers.
  • configure_subagent_models: a namespace-scoped model-facing tool for reading or updating this plugin's settings without filesystem access.
  • model-subagent-setup: a guided skill for selecting routes, generating routing guidance, obtaining confirmation, and saving through the constrained configuration tool.
  • Settings → Subagent Models: a Web settings page for manually adding, editing, and removing routes.
  • A hot-reloaded subagent-model-router namespace in ~/.dsh/settings.yaml.
  • Foreground execution and durable continuable background subagents.
  • An active-model chip in an opened subagent header.
  • Active-model chips in healthy rows of the parent session's subagent catalog.

With an empty models list, the catalog, configuration, and wait tools remain available alongside the settings page and setup skill, but subagent_model is not registered. This provides a bootstrap state for initial setup.

Orchestrator behavior

When model-selectable delegation is available, the system prompt tells the orchestrator not to duplicate work it delegated. After issuing all intended background delegations, it must call wait-for-subagents before synthesizing the child results or giving a final answer. The wait tool joins every continuable background child started by that parent, whether through a standard delegation tool or subagent_model; it remains useful when no model routes are configured. It preserves every terminal content block and drops retained records when the parent is disposed, while foreground calls already return their result directly. If another plugin already owns or scope-shadows the wait-for-subagents name, this plugin leaves it untouched and suppresses its wait-specific guidance and tracking for the affected agents.

Model identity chips

The opened subagent header and every healthy row in its parent's subagent catalog show the model id from the latest adapter-resolved request. Hover and accessible text expose the complete provider/model route. The plugin resets the route at the child's own descriptor so a fork cannot inherit its ancestor's model, and it omits the chip until the child records an authoritative request route.

Requirements

  • DeepSeek Harness 0.1.0-rc.6 or compatible
  • The Web profile and built-in subagent conversation UI
  • A preset exposing the normal skill loader/tool
  • The Host spawn subagent provider, included by standard DSH profiles

Install

# From npm once published
dsh plugin --profile web add dsh-subagent-model-router

# Or from inside this checkout
dsh plugin --profile web add .

Restart dsh web after installation and refresh the page. Open Settings → Subagent Models, or invoke:

/model-subagent-setup

Configure through the Web UI

The Subagent Models settings page provides controls for:

  • model alias, display name, LLM provider route, and exact model id;
  • comma-separated routing tags and the “when to use” description;
  • optional per-model output-token caps;
  • subagent backend, delegation depth, and background execution.

On DSH rc.6, the built-in Web settings API exposes only a fixed namespace allowlist. This plugin therefore uses a package-owned, same-origin Host endpoint backed by the same Settings service, schema validation, persistence, and revision conflict protection. Successful changes apply live: the old delegation tool is removed and the updated schema and prompt guidance are registered immediately.

The endpoint rejects non-loopback connections and cross-origin mutations, so the page is unavailable from a non-local browser connection.

Configure through the model-facing tool

configure_subagent_models is the preferred path for agent-assisted setup:

  • action: "get" reads the current normalized settings.
  • action: "update" replaces the complete model list and optionally changes the backend, depth, or background policy.
  • The tool calls the Settings service directly and can modify only subagent-model-router; it accepts no filesystem path and cannot read or write other namespaces.
  • Updates pass the same schema and provider-capability validation as the Web UI, persist to settings.yaml, and apply live.

The update action is intentionally documented for direct user-requested changes only. The setup skill must show the complete proposed list and receive explicit confirmation before calling it.

Configure through settings.yaml

Merge the following namespace into ~/.dsh/settings.yaml:

subagent-model-router:
  subagentProvider: spawn
  maxDepth: 3
  enableRunInBackground: true
  models:
    - alias: fast
      provider: acme
      model: acme-fast
      displayName: Acme Fast
      tags: [fast, routine]
      description: Use for quick, well-scoped tasks where low latency matters.
    - alias: deep
      provider: acme
      model: acme-reasoner
      displayName: Acme Reasoner
      tags: [reasoning, review]
      description: Use for difficult analysis, architecture decisions, and adversarial review.
      maxTokens: 16384

See examples/settings.yaml for a copyable document fragment. The settings file is watched; valid edits apply without changing a Cordis patch.

Settings reference

FieldDefaultPurpose
models[]Routes exposed to AI agents. An empty list leaves setup/catalog only.
models[].aliasrequiredStable selector shown in the tool's model enum.
models[].providerrequiredExact registered LLM provider route.
models[].modelrequiredExact model id interpreted by that provider.
models[].displayNamealiasHuman-readable label in routing guidance.
models[].tags[]Lowercase kebab-case routing tags.
models[].descriptionrequiredOne sentence describing when to use this route.
models[].maxTokensprovider defaultOptional cap for an initially created or resident child. DSH rc.6 does not restore it after a continuable child is cold-resumed.
subagentProviderspawnSubagent execution backend, not the LLM provider.
maxDepth3Maximum delegation depth enforced by the backend.
enableRunInBackgroundtrueEnable durable background children and default to them.

The delegation tool is always named subagent_model. Legacy toolName values from versions before 0.4 are ignored and are removed the next time the namespace is saved.

The live catalog is advisory: some adapters accept model ids they do not advertise. Manually entered ids remain allowed but should be user-confirmed.

Known limitations

DSH rc.6 has no additive slot inside a subagent catalog row, so the plugin owns the existing subagent-catalog header cell to render row chips. It claims that cell by registering at priority: -1 — a list cell renders its lowest live priority, and the host's own entry sits at the default 0. The subagent-model cell is registered the same way so a host build that also fills it stays shadowed rather than clashing. Catalog interaction changes in DSH must be mirrored here until the host exposes a row extension slot or renders subagentModelRoute itself.

Development

pnpm install
pnpm test

Project layout:

dsh-subagent-model-router/
├── lib/
│   ├── index.js
│   ├── client.js
│   └── model-catalog.js
├── skills/model-subagent-setup/SKILL.md
├── examples/settings.yaml
├── test/
│   ├── client.test.js
│   ├── model-catalog.test.js
│   └── plugin.test.js
├── cordis.patch.yml
└── package.json

The Client bundle is plain window.__ModuleLoader__.load(...) JavaScript and requires no build step.

License

MIT