DeepSeek Harness Plugin

toolclub/dsh-agent-team-gui

Stars ★ 1 Category Workflow & Automation Added 2026-08-15

Global persistent agent squads with per-agent model/tool policies: manage them in Settings, select and toggle one per conversation, then collaborate on normal sends in a fixed or model-planned order.

Install

# from GitHub (first run asks for allowBuilds approval — follow the hint, retry)

dsh plugin --profile web add github:toolclub/dsh-agent-team-gui

Any plugin you install runs third-party code with your own permissions — it can read your files, use your credentials, and reach the network, and tool approvals don’t sandbox it. GitHub-sourced plugins also run build scripts at install time. Only install sources you trust, and pin a commit (github:owner/repo#sha).

README

English | 简体中文

GitHub stars GitHub release MIT license

Persistent multi-model agent squads for DeepSeek Harness. Give every member its own model and tool policy, save reusable teams in Settings, then select a squad beside the composer and keep chatting normally.

Manage persistent multi-model squads in DeepSeek Harness Settings

60-second install

dsh plugin --profile web add -w github:toolclub/dsh-agent-team-gui#v0.1.0
dsh --profile web

Then open Settings → Teams, create a squad, select it beside the conversation composer, and turn on Squad mode. The plugin uses your existing dsh model routes and credential store; it never stores API keys.

What you get Why it matters
One provider/model and tool policy per agent Mix a strong planner, fast implementer, and strict reviewer in one squad
Global persistent agents and squads Build a team once, then reuse it across projects and conversations
Fixed or model-planned execution order Pin a repeatable workflow, or let the lead model plan roles for each task
Serial/parallel execution with spawn/fork/chain context Match collaboration topology to the task instead of forcing one pattern
Parent tool trace plus child sessions Inspect which member did what and diagnose failures

Enable a saved squad directly in a DeepSeek Harness conversation

If this workflow is useful, a GitHub Star helps other DeepSeek Harness users discover it. Bug reports and real squad recipes are equally welcome.

[!WARNING] Both dsh and this plugin are developer previews. The plugin targets dsh >=0.1.0-rc.5 <0.2.0 and requires Node.js 22.19 or later. dsh can make breaking changes before a stable release; pin both dsh and this plugin when reproducibility matters.

Key distinction

A squad does not have to share one model configuration. Every agent can independently select an existing dsh provider/model route, maxTokens, and tool allow/deny policy. Save those agents as global reusable definitions in Settings, combine them into persistent squads, and select a squad per conversation. When squad collaboration is enabled, an ordinary Send enters the collaboration flow: the squad follows its optional fixed member order, or lets the model plan assignments and execution order when no order is pinned.

Status and architecture

Settings --> global agent/squad definitions --+
Conversation --> per-session squad mode -------+--> dsh storage-domain --> JSON backend
                                              |
Conversation squad selector + collaboration toggle
                                              |
ordinary Send --> fixed member order, if set --+
              `-> model-planned roles/order otherwise
                                              |
                  Agent A / Agent B / Agent C
                                              |
             assistant response + traceable child sessions

Natural-language request --> dispatch_to_squad (model tool) --> same squad runtime

The package contains a Web client, loopback-only Connection RPC, host service, durable registry, and model-facing dispatch tool. Settings → Teams manages global definitions. Each conversation gets a squad selector and collaboration toggle; after selecting a squad and enabling collaboration, the user sends through the ordinary composer rather than a separate dispatch form. Provider and model choices are read from dsh's existing model configuration.

Current capabilities:

  • Durable agent/squad records and per-session squad-mode selection through dsh storage-domain.
  • Settings forms to create, edit, and delete global agents and squads, including a configured-model picker.
  • Per-conversation squad selection and an explicit collaboration toggle; when enabled, ordinary Send activates the selected squad mode and instructs the lead model to collaborate, while disabling it returns to normal single-agent sends.
  • Optional fixed member order on a squad; without one, the model determines member assignments and execution order from the request.
  • Per-agent { provider, model, maxTokens? } route and tool restrictions; no API-key storage.
  • Model-callable dispatch_to_squad with optional explicit per-agent assignments.
  • Serial or parallel execution and spawn, fork, or serial-only chain context modes.
  • Explicit per-member success/failure results. One member failure does not silently cancel the remaining squad.
  • Complete dispatch input/output in the parent session's append-only tool/call and tool/result records, with child session/run IDs in each member result.

Prerequisites

  • The dsh Web profile, version >=0.1.0-rc.5 <0.2.0. This bundle is not a headless or bare-profile bundle.
  • Node.js 22.19 or later.
  • pnpm on PATH. The Git-install notes below apply to pnpm 10 and later.
  • At least one provider/model route configured in dsh. Configure credentials through dsh Settings or its credentials mechanism, never in this plugin's records.

The commands below assume an installed dsh executable. Cloning the DeepSeek Harness source does not install that executable globally. From the Harness repository root, verify the source CLI with:

pnpm dsh --version

When running from another directory, replace every dsh ... below with pnpm --dir /absolute/path/to/deepseek-harness dsh ....

Install from a local directory

Run each command from the directory that contains dsh-agent-team-gui.

  1. Install the plugin's development dependencies:

    pnpm --dir ./dsh-agent-team-gui install
    

    Expected: pnpm finishes successfully and creates or updates dsh-agent-team-gui/node_modules.

  2. Build the checkout before linking it into a profile:

    pnpm --dir ./dsh-agent-team-gui run build
    

    Expected: the command exits with status 0 and produces the runtime entry under dsh-agent-team-gui/lib/.

  3. Add the local bundle to the Web profile:

    dsh plugin --profile web add -w ./dsh-agent-team-gui
    

    Expected: pnpm reports dsh-agent-team-gui as added. dsh must not print the warning that the package "declares no dsh.bundle".

  4. Inspect the composed configuration without booting it:

    dsh --profile web --dump-config
    

    Expected: output contains a dsh-agent-team-gui bundle layer and an agent-team-gui row.

  5. Start the profile:

    dsh --profile web
    

    Expected: dsh starts normally, Settings → Teams is available, and conversations show the squad selector and collaboration toggle. When info-level host logging is enabled, the log also contains [agent-team-gui] durable registry and dispatch_to_squad ready.

Install from a tarball

A built tarball contains compiled output and therefore needs no install-script allowance.

  1. Install dependencies and build:

    pnpm --dir ./dsh-agent-team-gui install
    pnpm --dir ./dsh-agent-team-gui run build
    

    Expected: both commands exit with status 0 and dsh-agent-team-gui/lib/ exists.

  2. Create the tarball:

    pnpm --dir ./dsh-agent-team-gui pack
    

    Expected: pnpm prints the generated archive name, normally dsh-agent-team-gui-0.1.0.tgz. Use the exact path printed by your pnpm version below.

  3. Install that archive:

    dsh plugin --profile web add -w ./dsh-agent-team-gui/dsh-agent-team-gui-0.1.0.tgz
    

    Expected: pnpm reports dsh-agent-team-gui as added without an allowBuilds prompt. If pack printed the archive elsewhere, substitute that exact path.

  4. Verify the layer:

    dsh --profile web --dump-config
    

    Expected: the dump contains the dsh-agent-team-gui layer and agent-team-gui row.

Install from GitHub

Git dependencies contain source rather than prebuilt release artifacts. This repository therefore ships a self-contained prepare path that builds its runtime entry without assuming a sibling dsh monorepo checkout.

You can also send this single sentence to a DeepSeek Harness agent that has terminal access:

Follow the README at https://github.com/toolclub/dsh-agent-team-gui and install the plugin into the DeepSeek Harness web profile; resolve and pin the current main commit SHA, configure pnpm allowBuilds as documented, and verify with dsh --profile web --dump-config.

  1. Pin and install a reviewed commit:

    dsh plugin --profile web add -w github:toolclub/dsh-agent-team-gui#<commit-sha>
    

    Expected with pnpm 10 or later on the first attempt: installation may fail because pnpm blocks the Git dependency's prepare script. pnpm prints the exact package key, and dsh prints the profile directory whose pnpm-workspace.yaml must be changed.

  2. Add exactly the key pnpm printed to that profile's workspace file. With the default dsh home, edit ~/.dsh/profiles/web/pnpm-workspace.yaml (or $DSH_HOME/profiles/web/pnpm-workspace.yaml when DSH_HOME is set):

    allowBuilds:
      dsh-agent-team-gui: true
    

    Expected: the YAML now preserves any existing workspace settings and contains the printed key under allowBuilds. Do not guess the key if pnpm printed a different one.

  3. Re-run the same pinned install:

    dsh plugin --profile web add -w github:toolclub/dsh-agent-team-gui#<commit-sha>
    

    Expected: pnpm is allowed to run prepare, builds the package, and reports it as added.

  4. Verify the bundle layer:

    dsh --profile web --dump-config
    

    Expected: the dump contains the dsh-agent-team-gui layer and agent-team-gui row.

  5. Restart any running dsh Web process, then refresh the browser. Installing or updating replaces files on disk but cannot replace an already loaded host module. The UI performs an RPC revision handshake and shows an explicit restart message if client and host revisions do not match.

[!CAUTION] allowBuilds authorizes that package to execute code on your machine during installation. This code runs outside every dsh agent sandbox. Allow only packages whose source you trust, review the selected revision, and pin github:owner/repo#<sha> so later pushes cannot silently change what executes. A built tarball avoids this build permission.

Configuration

The bundle inserts this host row:

- id: agent-team-gui
  name: dsh-agent-team-gui
  config:
    defaultProvider: spawn
    defaultExecutionMode: serial
    defaultContextMode: spawn
Field Type Default Meaning
defaultProvider string spawn Registered dsh subagent provider used unless dispatch/context selection chooses another one.
defaultExecutionMode serial | parallel serial Default member scheduling.
defaultContextMode spawn | fork | chain spawn spawn starts fresh children; fork includes the parent's completed-turn prefix; chain passes each serial member's text to the next.

To override it for one profile, edit $DSH_HOME/profiles/<name>/cordis.patch.yml:

- id: agent-team-gui
  config:
    defaultProvider: fork
    defaultExecutionMode: parallel
    defaultContextMode: fork

dsh patch rows replace the complete config object; they are not deep-merged. Restate every field you need whenever overriding the row. chain is valid only with serial execution.

Agent records contain:

Field Required Meaning
name yes Display name.
systemPrompt yes Role/persona passed to the child agent.
provider yes Existing dsh provider route name.
model yes Existing model id for that provider.
maxTokens no Per-agent token cap.
toolScope.allow / toolScope.deny no dsh tool-name restrictions applied to that child.

Squad records contain:

Field Required Meaning
name yes Global display name used by Settings and conversation selectors.
members yes Unique agent IDs available to the squad.
collabNote no Collaboration guidance included in member prompts.
executionOrder no Fixed complete ordering of every member. Omit it for lead-model planning.
executionMode no Squad default: serial or parallel; falls back to plugin config.
contextMode no Squad default: spawn, fork, or serial-only chain; falls back to plugin config.

Squad records contain name, an optional collaboration note, a member list, optional executionOrder, and optional executionMode/contextMode defaults. An agent may appear in multiple squads. Settings → Teams edits these global records through a loopback-only host RPC. A fixed executionOrder must contain every member exactly once. With no fixed order, the lead model plans assignments and passes a complete memberOrder when dispatching. The plugin stores route names only; it does not store or copy provider secrets.

In-process service API

Plugin authors can also use the same registry in-process:

const agentId = await ctx.agentTeamGui.createAgent({
  name: 'Reviewer',
  systemPrompt: 'Review for correctness and cite concrete evidence.',
  provider: 'your-configured-provider',
  model: 'your-configured-model',
  toolScope: { allow: ['bash', 'str_replace_editor'] },
})

const squad = await ctx.agentTeamGui.createSquad({
  name: 'Release review',
  collabNote: 'Run independent checks, then consolidate findings.',
  members: [agentId],
})

The service also exposes get/list/update/delete methods for both record types, addMemberToSquad, removeMemberFromSquad, exportDefinitions/importDefinitions, and a programmatic dispatch method. Exact TypeScript signatures are exported by the package declarations.

Usage examples

Natural-language dispatch (available after a squad exists)

User: Give the release-review squad this task: inspect the patch for regressions.
      Have the reviewer check correctness and the test agent run focused tests in parallel.

Assistant: [calls dispatch_to_squad with squadId, task,
            assignments=[...], executionMode="parallel", contextMode="spawn"]

Assistant: The squad completed with two member results. The reviewer found ..., and the focused
           tests .... Any failed member is listed explicitly instead of being omitted.

The model selects dispatch_to_squad; the plugin does not parse the user's text with regular expressions. squadId accepts either the durable ID or an exact squad name (case-insensitive); if names are duplicated, use the durable ID. Tool arguments are squadId, task, optional assignments: [{ agentId, task }], optional executionMode, and optional contextMode. memberOrder is also optional when the squad has no fixed executionOrder; when supplied, it must be a complete, duplicate-free ordering of all squad members. A fixed squad order cannot be overridden per call. The tool renders the complete canonical JSON result—including each member's runId, childId, status, error, stop reason, and output—so the lead model can produce the final summary.

Conversation collaboration toggle

  1. Start dsh --profile web and open Settings → Teams.
  2. Create the agents. Choose each provider/model from the routes already configured in dsh; optionally enter max tokens and comma-separated allowed/denied tools.
  3. Create a global squad, select its members, add an optional collaboration note, and optionally pin a fixed member order. Leave the order unset to let the model plan roles and ordering.
  4. Open any conversation, select Release review in its squad selector, and enable squad collaboration.
  5. Type the task in the normal composer and select Send. Disable the toggle to return that conversation to ordinary single-agent sends.
User types: Inspect this change and prepare a release recommendation.
Conversation control: Release review squad -> Collaboration on
User selects: Send

Assistant: [the selected squad collaborates using the current conversation as parent]
Assistant: The release-review squad recommends ... Reviewer: ... Test agent: ...

The selected squad is conversation-scoped and its mode is durable; both session modes and global agent/squad definitions survive restart. Deleting a selected squad automatically disables affected session modes. Sending does not require a second task box or a separate Dispatch button. Internally, a dynamic system prompt tells the lead model to call dispatch_to_squad once and summarize its result for the normal assistant response. This is a best-effort model instruction, not an API-level forced tool call; see Known limitations.

Export and import definitions

Settings → Teams can dump and restore the durable definitions as a JSON document.

  • Export downloads agent-team-gui-<date>.json containing { "format": "agent-team-gui/definitions", "version": 1, "agents": [...], "squads": [...] } — every record with its durable id, plus model routes (never API keys).
  • Import reads such a file and applies it with merge semantics: document rows are upserted by id, rows already in the store that the document does not mention are kept, and a squad may reference an agent that already exists in the store. The whole document is validated first — shape, duplicate ids, model routes, and squad member references — so a rejected import writes nothing. (The durable writes themselves are not a single transaction: a storage failure mid-import can leave a partial apply.)

The same operations are available in-process as exportDefinitions() and importDefinitions(document, mode); mode is merge (default) or replace. replace makes the document the entire store, and then a squad may only reference agents present in the document.

Observability and failure behavior

The parent session's ordinary tool/call and durable text tool/result events retain the request and complete canonical JSON aggregate. Every member result includes the provider-owned child session/run IDs when a child started, so its trajectory can be inspected through dsh's existing subagent/session views. The host log also records member start/finish/failure. Results distinguish complete, partial, and failed members. A member failure is aggregated with its error and does not silently stop unrelated members. The plugin contributes no long-lived subprocess; Cordis owns tool/listener cleanup, and the storage domain is closed when the plugin unloads.

Uninstall

The general form is dsh plugin --profile <name> remove <pkg>. For this Web bundle:

dsh plugin --profile web remove dsh-agent-team-gui

Expected: pnpm removes the dependency and dsh removes dsh-agent-team-gui from the profile's bundle list. Durable records under the dsh storage backend are not automatically deleted.

Known limitations

  • Web profile only; the bundle depends on dsh Web's storage, Connection RPC, and browser module services and does not support a headless or bare custom profile.
  • There is no separate shell CLI/YAML record editor; use Settings or the in-process service API.
  • No custom squad/* session event types: the current out-of-tree API cannot register them in dsh's known-event catalog. Observability relies on standard tool events, child sessions, and host logs.
  • The storage domain is version 0; developer-preview releases may reject or require migration of older on-disk data.
  • Model route names are validated by dsh when children run; a removed/misspelled provider or model produces an explicit member failure.
  • Squad mode is best-effort model orchestration: the dynamic system prompt instructs exactly one dispatch_to_squad call, but the current Harness generation API exposes no toolChoice control with which the plugin could hard-force that call.
  • Large fan-outs do not yet use the workflow engine's concurrency controls.
  • dsh APIs are pre-stable, so compatibility is intentionally bounded to >=0.1.0-rc.5 <0.2.0.

Roadmap

  • Add bulk editing and richer per-agent assignment controls in Settings.
  • Add schema migrations for durable definitions.
  • Add bounded concurrency and richer trajectory projections for large squads.

Contributing

  1. Open an issue describing the behavior and dsh version.
  2. Install dependencies with pnpm install and build with pnpm run build.
  3. Add focused tests and run pnpm test, pnpm run typecheck, and pnpm pack as applicable.
  4. Keep RPC loopback-scoped, never store API keys, and use only dsh APIs verified in the matching source version.
  5. Submit a focused pull request with an English commit message.

License

Released under the MIT License.

Content from the project README on GitHub ↗

Links

More in this category

View the whole category →