Ross ROSS = Recommend OSS · open-source software intelligence for agents

open-multi-agent/open-multi-agent

TypeScript AI agent orchestration framework with dynamic workflows. Describe the goal, not the graph: a coordinator plans the task DAG at runtime and runs it on any LLM (Claude, ChatGPT, Gemini, DeepSeek, or local models). observed · 2026-08-28

github.com/open-multi-agent/open-multi-agent · homepage · TypeScript · MIT (permissive) observed · 2026-08-28

Health v2 · maintenance only

81/100

  • Activity 99
  • Release rhythm 98
  • Longevity 11

Flags: young

How is this computed?

round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10) — computed 2026-09-03. Adoption (stars, forks) is never an input.

  • gap_med: 6
  • age_days: 155
  • days_rel: 12
  • days_push: 8
  • n_releases_24m: 22

Full methodology

Adoption not part of the score

6831 stars · 2427 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-29, confidence not recorded

Open Multi-Agent (OMA) is an MIT-licensed TypeScript framework for multi-agent AI orchestration that runs inside Node.js backends. A coordinator decomposes a goal into a task DAG at runtime, a deterministic scheduler executes tasks in parallel across a team of agents on any LLM provider (Claude, GPT, Gemini, DeepSeek, Ollama, or OpenAI-compatible endpoints), and every run is traceable, resumable, and replayable via an offline Run Viewer.

Use cases

  • orchestrate multiple AI agents on a single goal in a Node.js backend
  • decompose a goal into a parallel task DAG without hand-building the workflow graph
  • run a security review or PR review as a coordinated multi-agent team
  • mix models per agent, including local models via Ollama
  • include external coding agents (via ACP) as team members in one task DAG
  • trace, approve, budget, and replay multi-agent runs
  • get schema-validated structured output from agent tasks

When to choose

  • a goal benefits from multiple specialist roles and parallelizable work
  • you want runtime planning instead of maintaining a fixed workflow graph
  • you need orchestration embedded in an existing Node.js/TypeScript backend
  • you want provider-agnostic model access with tracing and resumability

When to avoid

  • one model call already solves the task
  • every step and transition must be fixed and deterministic
  • you only need a provider SDK or a chat UI
  • you want to avoid the extra cost of a planning model call

Facets

framework · maturity active

agent-framework llm-inference mcp workflow-automation sdk large-language-models developer-tools backend cross-platform self-hosted multi-agent agent-orchestration task-dag dynamic-workflows crewai-alternative langgraph-alternative openai anthropic claude gemini deepseek ollama local-llm opentelemetry acp run-viewer structured-output ai-agents automation nodejs typescript

9 sources

Member repositories

RepositoryRoleHealth v2
open-multi-agent/open-multi-agentmain81

For agents

markdown · JSON · MCP: product_card(name="open-multi-agent/open-multi-agent")

Data as of 2026-08-30T08:39:29.467469+00:00 · Report a problem