dsh-agent-teams
A DeepSeek Harness plugin that coordinates captain-led multi-agent teams with dependency-aware tasks, durable messaging, and a live web activity panel.
Install
$ dsh plugin --profile web add @nanmicoder/dsh-agent-teamsPlugin Overview & Capabilities
AI-assisted organization based on the public repository snapshot. The content must be grounded in source evidence and does not replace compatibility or security verification.
Key Capabilities
- Creates a captain-led team from the current DSH session.
- Adds durable, continuable sub-agent members with role-specific work.
- Splits goals into tasks with owners, explicit dependencies, and task states.
- Dispatches work through durable direct mailbox messages.
- Persists team state under <workspace>/.agent-teams/.
- Provides a web activity panel for team roles, work, messages, task dependencies, and archived history.
- Can snapshot the captain's active LLM provider, model, and reasoning effort for later member continuations.
Useful For
- Coordinate parallel reviews from performance, security, and product perspectives.
- Break a complex goal into dependent tasks assigned to specialized sub-agents.
- Run follow-up work with durable team members and persistent task context.
- Monitor multi-agent task progress in the DeepSeek Harness web UI.
Who It Fits
- DeepSeek Harness users coordinating multi-agent work.
- Developers and technical teams handling tasks that benefit from delegation and dependency tracking.
- Users who want persistent, in-session sub-agent collaboration without a separate workflow engine.
Documented Limitations
- Requires an existing DeepSeek Harness installation and the web profile for the documented installation command.
- One captain can lead only one active team at a time.
- Members act only when woken; messages remain in durable mail while participants are idle.
- State is serialized within one DSH process; concurrent processes editing the same team are not coordinated.
- The activity panel reports persisted state, and models may finish work without making the expected task-state update.
DSH Compatibility
Version-specific runtime evidence collected by DSH Plugin. A missing result means we have not tested that combination yet.
Security Signals
Objective signals discovered from package metadata and source inspection. These are not a guarantee that a plugin is safe.
package.json declares 0 runtime, 22 development, 14 peer, and 0 optional dependencies.
package.json declares DSH bundle metadata.
GitHub reports the repository license as MIT.
A root package.json was captured and can be inspected by the registry.
Public GitHub source metadata is available for this registry snapshot.
Source & Registry Notes
Public provenance, Registry classification, and the latest source check for this entry, kept separate from runtime verification.
- Source repository
- NanmiCoder/dsh-agent-teams
- Registry source
- GitHub · dsh-plugin topic
- Registry classification
- Plugin
- Source checked
- 2b11412 · 2026-08-15
This project is independently indexed from public source information. DSH Plugin is not affiliated with DeepSeek or the plugin author. Always check the author repository before installation.
Repository Activity
- GitHub stars
- 333GitHub stars
- Forks
- 30
- Open issues
- 7
- Last commit
- 2026-08-15
- Last release
- No release detected
Related DSH Plugins
Ranked by overlapping capabilities, use cases, plugin type, categories, and DSH profile.
Schedule self-contained coding tasks to run in fresh DeepSeek Harness Agent sessions.
View plugin →A DeepSeek Harness plugin for versioned, auditable continual evolution of prompts, memories, skills, and subagent specifications.
View plugin →An installable DeepSeek Harness bundle for robotics and embodied-intelligence workflows, spanning asset inspection, MuJoCo simulation, diagnostics, experiment evidence, and reports.
View plugin →Unofficial DeepSeek Harness LLM provider plugin for Command Code: live model catalog, reasoning-effort support, Models-page card. Ported from pi-commandcode-provider (MIT).
View plugin →