Multi Agent Orchestration

Use when coordinating genuinely independent Agent workstreams with task decomposition, dependency graphs, ownership, handoff contracts, approvals, recovery, and synthesis.

9 stars 3 forksAI, LLMs & Agents

Install

skills CLI (Claude Code, Cursor, Codex and more)
npx skills add sandbaseai/workbuddy-skill --skill multi-agent-orchestration
Manual install
git clone --depth 1 https://github.com/sandbaseai/workbuddy-skill.git /tmp/workbuddy-skill
mkdir -p ~/.claude/skills
cp -r /tmp/workbuddy-skill/skills/multi-agent-orchestration ~/.claude/skills/multi-agent-orchestration

Personal skills live in ~/.claude/skills. Use .claude/skills in a project folder to share a skill with your team.

From the SKILL.md

# Multi-Agent Orchestration

Use multiple Agents only when specialization or safe parallelism outweighs
coordination cost. Do not delegate a small, tightly coupled, or inherently
sequential task merely to increase Agent count. Delegation never expands the
requester's authority.

## Plan contract

Capture the measurable objective, completion criteria, explicit non-goals, task
graph, available Agents/tools, concurrency limits, shared resources, authority
boundaries, deadline, budget, and final decision owner. Produce a directed
acyclic graph with one accountable owner per task, dependencies, inputs, output
contract, write scope, validation, timeout, retry bound, escalation route,
approval point, and stop condition.

## Design the workflow

1. Decompose by separable outputs, not vague roles; keep shared mutable state
   minimal and tightly coupled steps under one owner.
2. Identify the critical path. Parallelize only tasks with independent inputs
   and non-overlapping side effects.
3. Give each Agent the minimum context and permissions needed. Include source
   artifacts rather than hidden conclusions when independent judgment matters.
4. Require structured handoffs containing status, result, evidence, changed
   state, assumptions, risks, and next dependency. Acknowledge receipt before
   downstream mutation.
5. Monitor dependency state and useful progress. Bound retries and debates;
   never recursively delegate without explicit capacity and ownership.
Read the full skill on GitHub

Details

Skill name
multi-agent-orchestration
Path
skills/multi-agent-orchestration/SKILL.md
License
MIT
Repo updated
8 September 2026

FAQ

What does the Multi Agent Orchestration skill do?

Use when coordinating genuinely independent Agent workstreams with task decomposition, dependency graphs, ownership, handoff contracts, approvals, recovery, and synthesis.

How do I install the Multi Agent Orchestration skill?

skills CLI (Claude Code, Cursor, Codex and more): npx skills add sandbaseai/workbuddy-skill --skill multi-agent-orchestration. Skills go in ~/.claude/skills (personal) or .claude/skills (project), and Claude loads them automatically when a task matches the description.

Is Multi Agent Orchestration popular?

It lives in sandbaseai/workbuddy-skill, which has 9 GitHub stars and 3 forks.

Related skills

Browse the most popular Claude skills, all ai, llms & agents skills, or write your own with the skills generator.