AgentIndex icon
AgentIndex
ToolsMCP ServersHOTCategoriesTrendingNewBest ListsCompareGuides
Submit Tool
ToolsMCP ServersHOTCategoriesTrendingNewBest ListsCompareGuides
Home/
Multi-Agent/
agency-orchestrator
agency-orchestrator logo

agency-orchestrator

Active·★ 2.4k·Apache-2.0·Updated 2026-10-22

Agency Orchestrator turns a one-sentence goal into a multi-step workflow run by several AI roles. `ao compose "..." --run` asks a model to break the task into steps, matches each step to a role from a library of 179 prebuilt roles, runs them in parallel where the steps allow (a DAG), and combines the output. The README's demos show a startup plan produced by five roles in about three minutes and a short story written by four roles. Workflows can also be written by hand in YAML, with built-in templates, a `validate` command, a plan view and a resume option for iterating.

Install with `npm install -g agency-orchestrator`. It supports nine LLM providers, six of which need no API key because they reuse a CLI you already pay for or run locally: the Claude Code, Gemini, Copilot and Codex CLIs, OpenClaw and Ollama. DeepSeek, Claude and OpenAI work with API keys, and custom `base_url` values allow other OpenAI-compatible services. The README also lists integrations with 14 AI coding tools and a programmatic API. The README is written in Chinese with an English version linked, and the project is Apache-2.0 licensed.

This write-up is based on the project's README and public repository data. We have not hand-tested the tool. Data as of 2026-10-18.

License
Apache-2.0
Latest release
desktop-v0.4.12 · 2026-10-08
Last commit
2026-10-18
Contributors
3
Forks
314
Open issues and PRs
9
First published
2026-03-21
npm package
agency-orchestrator 0.20.0 · 225 downloads last week

Source: GitHub, npm and PyPI public APIs.

#agent-orchestration#ai-agents#aider#autogen-alternative#claude#codex#copilot#crewai-alternative
$ Install
$ npm install -g agency-orchestrator
↗ Visit site★ GitHub
01

Features

01One-sentence task decomposition and multi-role orchestration
02179 specialized AI roles for diverse expertise
03YAML-based zero-code workflow definition
049 LLM providers with 6 requiring no API key
05Automatic DAG parallelism and resume capability
02

Why choose it

+One sentence is enough to get a multi-role workflow, with no YAML needed to start
+179 prebuilt roles and built-in workflow templates
+Six providers need no API key, reusing Claude Code, Gemini, Copilot or Codex CLI access or local Ollama
+Parallel execution of independent steps, with validate, plan and resume commands
03

Trade-offs

!The demo outputs (business plans, revenue figures) are generated by models and are not verified, so treat them as drafts
!Several roles and steps multiply token use; the README's demos report 6k to 65k tokens per run
!Using subscription CLIs this way depends on those providers' terms for automated use
!The README is mainly in Chinese, with the English version in a separate file
04

Maintenance check

ActiveChecked: 2026-10-22
Last commit
2026-10-22
Commits, last 90 days
100+
Releases, last 90 days
12
Latest release
desktop-v0.4.14 (2026-10-21)
Open issues
5
Recent issues with at least one comment
3/6
Commits per week, last 12 weeks
Commits per week, last 12 weeks: 19, 42, 66, 117, 13, 8, 25, 79, 5, 9, 8, 33

Oldest week on the left; the latest four weeks are orange. Source: GitHub commit activity.

Status rule: archived on GitHub = Archived; last commit within 90 days = Active; 91 to 365 days = Slowing; older = Stale. All figures come from the GitHub API on the date shown.

Most discussed open issues

  • #91 Studio：受控读取本地项目并执行计划（原画布保存问题已修）4 comments · 2026-07-09
  • #104 Possible complementary direction: LongHorizon-Harness for sustained agent tasks2 comments · 2026-08-12
  • #66 桌面端发布完善：macOS 公证 + Windows 签名（打包/Windows 已完成）2 comments · 2026-06-19
  • #44 桌面端：目录设置 UI + 角色资源按需安装（文档已完善）2 comments · 2026-06-17
  • #194 [Bug] dsh-cli 桌面版静默空输出：引擎子进程用 Electron 内置 Node 20，而 dsh >=0.2 实际需要 Node >= 24.20 comments · 2026-10-17

Issue titles are quoted from GitHub. Open the link to check whether a problem still applies to the version you use.

05

Compatibility

claude-code
Claude Code CLI
Verified via docs
gemini-cli
Gemini CLI
Verified via docs
copilot-cli
Copilot CLI
Verified via docs
codex-cli
Codex CLI
Verified via docs
openclaw-cli
OpenClaw CLI
Verified via docs
ollama
Ollama
Verified via docs
06

Quick start

1
$ npm install -g agency-orchestrator
07

Use cases

↳Business plan generation (e.g., TikTok e-commerce opportunity analysis)
↳Product requirement review with cross-functional AI experts
↳Collaborative novel writing with narrative, psychology, and design roles
08

Alternatives

⚡Top VS:
VSagent-fleet-o★76→VSagent-network★101→VSClaudeR★354→VSclaude-plugins★97→
agent-fleet-o logo
agent-fleet-o★ 76
Open-source AI agent orchestration platform — self-hosted mission control for autonomous multi-agent systems. Visual DAG workflows, 450+ MCP tools, human-in-the-loop approvals. Works with Claude, GPT-4o, Gemini, Ollama, Codex.
VSCompare→
agent-network logo
agent-network★ 101
多 Agent，一行命令。Run a multi-agent network with Claude / GPT / MiniMax — one npm install.
VSCompare→
ClaudeR logo
ClaudeR★ 354
Connect RStudio to Claude Code, Codex, Gemini, and other LLM agents via MCP. Multi-agent orchestration, automated manuscript auditing, and zero-config setup with uvx
VSCompare→
claude-plugins logo
claude-plugins★ 97
28 plugins and MCP servers for Claude Code — TDD, multi-agent orchestration, iterative refinement, binary RE, structured decisions. Install any skill in one command.
VSCompare→
AI-Agents-Orchestrator logo
AI-Agents-Orchestrator★ 96
🪈 Intelligent orchestration system that coordinates multiple AI coding assistants (Claude, Codex, Gemini CLI, Copilot CLI) to collaborate on complex software development tasks via REPL or a Vue/Nuxt UI dashboard. Also includes an Agentic Team runtime with role-based multi-agent open communication & lead-gated final responses.
VSCompare→
better-code-review-graph logo
better-code-review-graph★ 76
Knowledge graph for token-efficient code reviews — fixed search, qualified call resolution, dual-mode embedding (ONNX + multi-provider cloud)
VSCompare→
Auto-claude-code-research-in-sleep logo
Auto-claude-code-research-in-sleep★ 17.9k
ARIS ⚔️ (Auto-Research-In-Sleep) — Claude Code skills for autonomous ML research: cross-model review loops, idea discovery, and experiment automation via Codex MCP
VSCompare→
zabbix-mcp-server logo
zabbix-mcp-server★ 244
MCP server for the complete Zabbix API — 220 tools, multi-server support, bearer auth, systemd ready. Works with ChatGPT, Claude, VS Code, Codex, JetBrains and any MCP client.
VSCompare→
See all alternatives →

Comments

Log in to leave a comment

No comments yet. Be the first!

On this page
01Features02Why choose it03Trade-offs04Maintenance check05Compatibility06Quick start07Use cases08Alternatives
jnMetaCode /
agency-orchestrator
↗

Describe a goal in one sentence and a team of AI roles collaborates on a plan

2.4k314Apache-2.0
Last commitToday
StatusActive
CategoryMulti-Agent
Contributors3
Latest releasedesktop-v0.4.12
Links
Documentation↗Discussion↗Issues↗Releases↗
Cloud Infrastructure
Ad

Deploy on DigitalOcean

Host your MCP servers, Docker containers, and AI agents with $200 free credit over 60 days.

Claim $200 Free Credit
© 2026 AgentIndex.app|Built by a 10-year iOS Developer.
Best by use caseAboutContactAdvertisePrivacy PolicyTerms of UseQYSGitHubBuy me a coffee ☕

Browse by Category

MCP ServersCode AssistantWorkflow AutomationRAG / Knowledge BaseMulti-AgentBrowser AutomationLLM InfraDev ToolingObservability

Not affiliated with Anthropic, OpenAI or Microsoft.