# NanmiCoder/dsh-agent-teams

AgentTeams plugin for DeepSeek Harness

Repository: https://github.com/NanmiCoder/dsh-agent-teams
Canonical: https://ross.abutalabs.com/products/dsh-agent-teams
Language: TypeScript
License: MIT
License Family: permissive
Topics: agentteams, dsh, deepseekharness, dsh-agent-teams, dsh-plugin
Last push: 2026-08-26T10:04:26+00:00

## Health v2 (maintenance only)
Score: 79/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 99, release rhythm 99, longevity 1
- inputs: {"age_days": 21, "days_push": 7, "days_rel": 11, "gap_med": 0, "n_releases_24m": 4}
- flags: young
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1067, forks 85 (observed 2026-08-28T04:03:24.776824+00:00)

## What it is
A DeepSeek Harness plugin that turns a single session into a captain coordinating durable sub-agents as a team. It provides dependency-aware task management, direct messaging, persistent state, an automatic scheduler, and a live Web UI.

## Use cases
- coordinate multiple AI agents on one goal
- split a complex task into dependency-aware subtasks
- delegate work to persistent sub-agents from one prompt
- run a multi-agent team inside DeepSeek Harness
- monitor agent team activity in a live web UI
- recover and retry stranded agent tasks automatically

## When to choose
- you use DeepSeek Harness and want multi-agent coordination without a separate workflow engine
- you need durable, continuable sub-agents with task dependencies
- you want a built-in web UI to watch team progress

## When to avoid
- you don't use DeepSeek Harness
- you need a standalone multi-agent framework independent of DSH
- you only need simple single-agent prompting

## Facets
- artifact type: plugin
- maturity: active
- function: agent-framework, workflow-automation, scheduling
- domain: large-language-models, developer-tools
- platform: cli
- tags: multi-agent, deepseek-harness, dsh-plugin, agent-teams, task-coordination, sub-agents, typescript, ai-agents, automation, nodejs, web-server

## Member repositories
- NanmiCoder/dsh-agent-teams (main) score 79

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:03:24.776824+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T06:55:09.084642+00:00, confidence not recorded.
  - readme: https://github.com/NanmiCoder/dsh-agent-teams (fetched 2026-08-28T04:03:24.776824+00:00, sha d3c30d4b810e)
- Data as of 2026-08-30T08:39:29.467469+00:00.