Framework

CHORUS The Multi-Agent Organization

CHORUS is a six-skill coordination protocol for teams of 2 to 15 people who share work with specialised AI workers: Charter, Handoff, Orchestrate, Review, Update, Sync. Org charts name reporting lines. CHORUS names coordination lines.

The moves
6

When to use it

Use CHORUS when adding agents made the team slower: lost handoffs, unreviewed model output, standups that could have been a dashboard. Protocol outlives tools.

The moves

  1. 01

    Charter

    Seven fields that turn a vague intention into a contract any worker, human or agent, can be handed.

  2. 02

    Handoff

    A context packet with six fields that carries a task across a boundary so the receiver never guesses.

  3. 03

    Orchestrate

    Serial steps, fan-out to parallel agents, fan-in to a merge, and a named escalation to a human.

  4. 04

    Review

    A gate where output passes only when it meets acceptance criteria, instead of being vibe-checked.

  5. 05

    Update

    Review findings flow back into prompts, tools, SOPs, and charters, so the system improves run over run.

  6. 06

    Sync

    One surface showing who owns what and what is blocked, read in a short weekly ritual.

The book

Available now

The Multi-Agent Organization

How Humans and AI Agents Share Work Without Losing the Plot

Open the book
The article The series

Related frameworks

STACK

The Agentic Codebase

STACK is a five-layer architecture for repositories where humans and agents share work: Structure, Toolchain, Agent configuration, Connection, Knowledge and quality. The agent OS is versioned like the app.

SENSE

AI Agents for Startup Strategy

SENSE is a five-move strategic intelligence layer: Scan, Evaluate, Navigate, Simulate, Execute. It turns agent work from a productivity shortcut into a sourced claim that changes a watchlist, an assumption register, or a Monday decision.

GRAIN

The RAG Engineer

GRAIN is a retrieval loop: Gather, Rank, Assemble, Inspect, Navigate. Gather governs the corpus: sources, permissions, and versioning. Rank retrieves candidate passages and reranks them for the question actually asked. Assemble builds the context the model may see. Inspect traces a wrong answer back to the chunk and retrieval decision behind it. Navigate keeps the index fresh: TTL policy, re-ingest, and change detection. If you cannot point at the passage, you do not have retrieval. You have a fluent guess.

Terminology

CHORUS here is the coordination protocol from The Multi-Agent Organization: Charter, Handoff, Orchestrate, Review, Update, Sync, for human teams of 2 to 15 with specialised AI workers. It is not Chorus AIDLC, not chorusprotocol.dev, not a choir, not CrewAI, and not a graph of model calls. ROSTER is the company-of-one counterpart.

Citation

CHORUS (The Multi-Agent Organization). Len P. van der Hof, Systems for the Strategic Self. https://lenvanderhof.com/frameworks/chorus/

Markdown for LLMs