DeepSeek Harness Plugin Hub

Publish and manage complete Harness Profiles. Discover Plugins for your next setup.

Explore

PluginsPresetsDocsNews

Community

Publish a pluginContactReport an issue

Resources

Plugin Hub on GitHubDeepSeek HarnessSystem statusPrivacy notice
© 2026 DeepSeek Harness Plugin HubPowered byPaxTech

Independent and unofficial. Not affiliated with, authorized by, or endorsed by DeepSeek.

Agent Teams Optimizer — DSH Plugin for DeepSeek Harness
← Plugins
A

dsh-agent-teams-optimizer

Agent Teams Optimizer

Configure the real Agent Teams teammate limit in DeepSeek Harness

The plugin will be installed here. Keep web if you are unsure.

npx -y @deepseek-ai/dsh plugin --profile web add github:Soulize/dsh-agent-teams-optimizer#1e594c79c52dd970505dbc6ca1896f0af978302f
READMECompatibilityVersions

Compatibility and provenance

Agent Teams Optimizer is published as dsh-agent-teams-optimizer and currently resolves to version 0.1.0. The Hub verifies its manifest and preserves the exact installation source for reproducible installs.

DSH compatibility
*
Runtime surfaces
web
Release source
github
Registry updated
10/19/2026

Versions

0.1.0stable
10/19/2026

Related plugins

Loading related plugins…

Latest
0.1.0
DSH
*
HMR
Process restart
Tree shaking
Safe tree shaking not declared
Unpacked size
Unavailable
Files
Unavailable
Surface
web
License
MIT
Source
github
GitHub
★ 1
Weekly downloads
0
Last push
10/19/2026
View source ↗
README badge

Click the badge to copy Markdown for your README.

Do you maintain this Plugin?Claim benefit · Priority security scan

Verify the GitHub repository declared in package.json to manage this listing. After you claim it, Hub will prioritize a security scan of the current version and publish the result when it passes.

Claim this Plugin →
Report an issue
DeepSeek Harness Plugin Hub
ProfilesPluginsCategoriesNewsDocsSign inManage Profiles
ProfilesPluginsCategoriesNewsDocsSign in

Related plugins

More verified plugins in agents-orchestration.

Headless@deepseek-ai/dsh-headlessThe dsh one-shot bundle: a direct core Agent/Session runner over dsh-base with no Host, HTTP, or browser layerExperimental Agent Team Profile@deepseek-ai/dsh-experimental-agent-team-profileAgent Teams collaboration, tools, and Web UI in one experimental bundleExperimental Ralph Bundle@deepseek-ai/dsh-experimental-ralph-bundleRun explicitly requested repeated work with fresh agents.Subagent Codex@deepseek-ai/dsh-subagent-codexOne-shot Codex subagent provider over the official app-server protocol

README

dsh-agent-teams-optimizer

An independent DeepSeek Harness plugin to configure the actual Agent Teams teammate admission limit. It changes the number checked when spawn_teammate creates a new, durable teammate; it does not merely ask the model to use more teammates.

In the official Agent Teams profile, agent-team is configured with maxMembers: 8. The underlying TeamService default is 16 when no profile override is provided. This plugin defaults to 16 and allows any positive safe integer.

Install

First enable the official Agent Teams bundle. Then install the optimizer in the same profile:

dsh plugin --profile web add "github:Soulize/dsh-agent-teams-optimizer#main"

Restart DSH after adding the plugin. In Web/Desktop, open Plugins -> dsh-agent-teams-optimizer -> agent-teams-optimizer -> Configure, enter the desired maxMembers, and save. DSH stores the option in the current profile; changes are applied to the active Team service without needing to restart, provided the current official internal layout is compatible.

For direct profile configuration, set:

- id: agent-teams-optimizer
  config:
    maxMembers: 32

The plugin's own bundle patch only inserts agent-teams-optimizer. It intentionally does not overwrite the official Agent Teams config row or any unrelated settings.

Behavior

  • The limit counts teammates only; the Lead is not counted.
  • It is a cumulative, immutable roster limit, not a simultaneous-running limit. Even an inactive or failed member keeps its slot; increasing the limit permits more distinct names. Lowering the limit does not delete historical members.
  • It changes the real TeamRoster admission field, so existing and future teams in the current Team service honor the new value.
  • Regular Subagent, workflow and nested delegation limits are not changed.
  • If Agent Teams is disabled, the optimizer remains idle. It binds automatically when the official agentTeams service becomes available.
  • Disabling/unmounting this plugin restores the original roster limit only if no other component has changed the same field.

Implementation and compatibility

The official TeamService currently captures maxMembers in its TeamRoster constructor and offers no public runtime setter. Modifying the agent-team Cordis profile row is the supported method for setting the limit at service startup. This plugin instead provides an editable, live UI by conditionally updating the existing roster's maxMembers property. This is an undocumented internal field, not a stable DSH API.

The adapter checks that the official service and writable field have the expected shapes before changing anything. If a future DSH release changes those internals, the plugin logs a warning and does not pretend the setting took effect. It does not replace spawnTeammate, monkey-patch tool handlers, or mutate the roster's historical records.

For a stable configuration-only alternative, use this profile override (the whole config object is replaced by a patch, so include the other fields to retain the official explicit values):

- id: agent-team
  config:
    maxMembers: 32
    maxTasks: 256
    maxMessageBytes: 65536
    disposalTimeoutMs: 5000

These two approaches should not be used to compete for the same limit at runtime. If both are present, the optimizer treats the effective value seen on service mounting as its restoration baseline.

Verify

When a compatible Team service mounts, check the DSH Host log for a line such as:

agent-teams-optimizer: maxMembers=32 (previous=8, official baseline=8, reason=service-mounted)

Test by creating more than eight distinct teammates in one Team. A ninth successful creation confirms that the runtime limit, not merely the prompt, has changed. Actual model behavior is a separate matter: raising the cap does not force the Lead to spawn more agents.

Run the unit tests with npm test (uses Node's built-in test runner).

Requirements

  • Official DeepSeek Harness Agent Teams service (@deepseek-ai/dsh-experimental-agent-team), not a third-party Teams implementation.
  • DeepSeek Harness >=0.2.0-rc.1; no maximum DSH version is imposed. Compatibility is verified by the runtime adapter on each mount.
  • Node.js >=22.19.0 for DSH.

License: MIT.