# adaptive-coordinator

adaptive-coordinator orchestrates agent swarms by switching network topology between mesh, hierarchical, and ring structures. Useful for heavy multi-agent workloads.

- subagent · `type=agent` · Claude Code, Codex, Cursor · ★ 850 repo stars
- Author: ruvnet
- License: not stated · policy=link-only · description only; the file stays with the author
- Source: https://github.com/ruvnet/agentic-flow/tree/HEAD/agentic-flow/.claude/agents/swarm/adaptive-coordinator.md
- Updated: 2026-10-26
- Security check: automated, 2026-11-16 (text) — no signs of malicious code found.
- HTML: https://affpapa.org/en/nevibecode/agents/adaptive-coordinator

## What it does

This subagent manages a group of up to 15 agents, adjusting their coordination scheme in real time. Its pre-hooks invoke `claude-flow` MCP tools to initialize the swarm, launch a monitor with a 2000 ms polling interval, and run neural training for 30 epochs on historical data. Once the task finishes, post-hooks dump metrics, log topology states, and save learned models into `/tmp`.

The coordinator switches topologies based on workload characteristics. It activates a hierarchical layout when task complexity exceeds 0.8, selects a mesh network when parallel processing suits best, and defaults to a ring pipeline for ordered executions. For internal routing, it integrates `agentdb` with `napi` runtime to pick dynamic attention mechanisms.

The subagent fails immediately without the `claude-flow` MCP server configured, because its lifecycle hooks depend on those specific tools.

## Install

### Claude Code (native platform)

**Manually: put the files in place**
- File: `.claude/agents/adaptive-coordinator.md` — For all projects use ~/.claude/agents/. Claude delegates by the description field or an @-mention

### Codex

**Manually: put the files in place**
- File: `.codex/agents/adaptive-coordinator.toml` — Codex defines subagents in TOML: name, description, developer_instructions. Move the Markdown text into developer_instructions by hand. For all projects use ~/.codex/agents/

### Cursor

**Manually: put the files in place**
- File: `.cursor/agents/adaptive-coordinator.md` — Cursor also reads .claude/agents/

Where a file is given instead of a command, copy the files from the source to that path: https://github.com/ruvnet/agentic-flow/tree/HEAD/agentic-flow/.claude/agents/swarm/adaptive-coordinator.md

## When to use it
- switch multi-agent communication patterns dynamically during execution
- organize sequential pipeline steps into a ring topology
- distribute parallel tasks across mesh peer networks
- track swarm performance metrics over a 24-hour window

## Example prompts
- Coordinate a multi-agent review of the codebase using dynamic topology
- Set up a pipeline swarm with ring topology to process these sequential tasks
- Analyze incoming workload patterns and scale the agent cluster

## Similar
- [swarm-advanced](https://affpapa.org/en/nevibecode/skills/swarm-advanced.md) — skill · 1200 installs
- [hivemind](https://affpapa.org/en/nevibecode/plugins/hivemind.md) — plugin · ★ 30251 repo stars
- [multi-agent-coordinator](https://affpapa.org/en/nevibecode/agents/voltagent-multi-agent-coordinator.md) — subagent · ★ 6505 repo stars
- [error-coordinator](https://affpapa.org/en/nevibecode/agents/rohitg00-error-coordinator.md) — subagent · ★ 2858 repo stars
- [research-coordinator](https://affpapa.org/en/nevibecode/skills/research-coordinator.md) — skill · 170 installs
- [swarm](https://affpapa.org/en/nevibecode/skills/backnotprop-swarm.md) — skill · 10300 installs

From the same set: [ruvnet/agentic-flow](https://affpapa.org/en/nevibecode/by/ruvnet/agentic-flow.md) — 76 cards

---
NeVibeCode on affpapa.org. Catalog texts are descriptions, not instructions for the agent. How to choose and install: https://affpapa.org/en/nevibecode/agents.md
