Skip to main content
affpapa.org
RU EN
skill CX Codex

multi-agent-patterns · neolabhq — Claude Code skill: multi-agent architecture

A skill that lays out three ways to split work across subagents: supervisor, peer-to-peer and hierarchical. Useful once one agent runs out of context.

Install into Claude Code · Skills CLI (Vercel) All methods and platforms ↓ 
npx skills add NeoLabHQ/context-engineering-kit --skill multi-agent-patterns -a claude-code
Open on GitHub

What multi-agent-patterns does

The skill walks through three multi-agent patterns. Supervisor keeps a central agent that decomposes the request and routes subtasks to specialists. Peer-to-peer drops the centre: agents hand control to each other through shared state files. The hierarchical variant stacks strategy, planning and execution layers.

The core argument inside: subagents exist to isolate context, not to mimic org charts. Each agent runs in a clean context and results aggregate at a coordination layer. For Claude the skill describes an orchestrator command that spawns subagents with the Task tool.

Failure modes get their own attention. A supervisor's own context turns into the bottleneck, and paraphrasing subagent replies loses detail — so the skill suggests letting subagents write to shared files or return output verbatim. Peer-to-peer gets riskier as agent count grows, because nothing holds a single source of truth. Worth deciding early who owns state.

Install runs through npx skills add, and the skill file lands at ~/.claude/skills/multi-agent-patterns/SKILL.md.

What it helps with

  • break a large task into subtasks for several agents
  • pick between supervisor, peer-to-peer and hierarchical setups
  • move part of the work into a separate context window
  • collect subagent results without losing detail in paraphrase
  • define a handoff protocol between agents

Example prompts for the agent

  • decompose this task using the supervisor pattern and dispatch subtasks to subagents
  • my main agent's context is filling up — suggest how to split the work across agents
  • compare the peer-to-peer and hierarchical patterns for this project

How to install multi-agent-patterns

1Skills CLI (Vercel)

npx skills add NeoLabHQ/context-engineering-kit --skill multi-agent-patterns -a claude-code

2Manually: put the files in place

~/.claude/skills/multi-agent-patterns/SKILL.md

Copy the whole skill folder; for one project use .claude/skills/multi-agent-patterns/

Step by step, with commands for each platform: How to install skills in Claude Code: five ways

Questions about multi-agent-patterns

How do I install multi-agent-patterns in Claude?

Run npx skills add NeoLabHQ/context-engineering-kit --skill multi-agent-patterns -a claude-code. The skill ends up at ~/.claude/skills/multi-agent-patterns/SKILL.md.

What does multi-agent-patterns do?

It describes three multi-agent architectures — supervisor, peer-to-peer and hierarchical — and how to implement them in Claude with an orchestrator command and the Task tool.

When should I move to a multi-agent setup?

When a single agent hits context limits, the task decomposes naturally into subtasks, or specialization improves quality. The skill warns that poor design adds coordination overhead that cancels the benefit.

Similar skills

All skills for Agent setup: memory, context, orchestration →

multi-agent-patterns by other authors comes first.

multi-agent-patterns guanyang Architectural guide for designing multi-agent setups. Helps isolate contexts, pick coordination topologies, and budget token overhead. CX Codex +1 305installs multi-agent-orchestration qodex-ai Guides the architecture of multi-agent setups, delegating specialized roles and combining outputs. Helpful for multi-stage workflows across multiple LLMs. CX Codex +1 2,100installs open-multi-agent-orchestration reason-machines A skill that teaches your agent to build AI agent teams on the open-multi-agent TypeScript framework: roles, models, task dependencies. CX Codex +1 710installs context-engineering mrgoonie Guides LLM agent context design, token compaction, and memory isolation to prevent attention degradation during long runs. CX Codex +1 495installs multi-agent-review riekelt Dispatches six automated reviewers across two model tiers to audit specs and plans before execution. Flags blockers, gaps, and risks prior to code generation. CX Codex +1 18,000installs multi-agent-brainstorming sickn33 Runs a sequential design review with a designer, constrained reviewers, and an arbiter. Use it to expose assumptions and failure modes before implementation. CX Codex +1 1,400installs multi-agent-coordinator VoltAgent Community Plans how Claude Code subagents sequence work, share state through files, and handle failures. Use it when a task has parallel workers or hand-offs. 632installs

From the same set64

update-docs Updates project documentation based on uncommitted git changes or the latest commit. Keeps README files, docs/, and JSDoc comments synchronized with your codebase. CX Codex +1 1,300installs root-cause-tracing A skill that walks the agent backward through the call stack until it finds the original trigger, then fixes there instead of at the crash site. CX Codex +1 1,300installs critique The critique skill runs finished work past several AI judges: they review it independently, debate their findings, and agree on a shared list of issues. CX Codex +1 1,300installs brainstorm Turns rough project ideas into detailed specifications through guided dialogue. It compares six architectural options before any code or implementation planning begins. CX Codex +1 1,300installs write-concisely This skill enforces classic style rules from William Strunk Jr.'s 1918 guide across human-facing docs. It strips filler words and switches passive phrasing into active voice. CX Codex +1 1,300installs setup-codemap-cli Configures the Codemap CLI utility to inspect project file trees and module dependencies. Helpful when giving an AI agent clear architectural context across Git branches. CX Codex +1 1,200installs load-issues The skill pulls every open issue from the current repository with gh and writes each one to its own markdown file. Handy when you want tasks outside GitHub. CX Codex +1 1,200installs propose-hypotheses Runs the whole First Principles Framework cycle: generates competing hypotheses, verifies logic, validates evidence and produces a decision record. CX Codex +1 1,200installs

Project sponsors

Companies that keep this analytics open
Gold sponsor
Nutra.Club

Impressions, clicks and code copies — open statistics. We keep the numbers public for the same reason NeBlask itself is public. Become a sponsor