Skip to content
Back/Harness Engineering

Shared Brain First, Boundaries Second

View in Graph
Updated 2026-10-31
4 min read
762 words

Shared Brain First, Boundaries Second

What it is

A design principle for multi-agent systems: before adding agents, build the shared memory layer first. The shared brain is not a giant bucket of merged memories, but a curated layer of team-level facts that persist across agents. Individual agents keep private notebooks for personal habits and temporary context.

Why it matters

When multiple agent platforms run in parallel, each agent remembers decisions independently but remains unaware of the other's memory. This leads to repeated questions, duplicated decisions, and fragmented context. The shared brain prevents agents from "acting like strangers sharing a house."

Every 的公司级重建(2026-10)在更大尺度上复现了同一判断:他们放弃了"每人一个 Agent"的舰队,改为全公司共享一个 Agent,核心理由正是个人 setup 里的好工作流无法复利——增长负责人 Austin Tedesco 的个人 Agent 接上分析工具后,全团队都去问他的 Agent,共享形态是这种"引力中心"的制度化。

Key points

  • Shared study vs private notebook: The shared study stands on its own. It holds long-term decisions, stable facts, important relationships, and context that agents repeatedly reach for. Private notebooks contain personal habits, temporary context, and role-specific experience.
  • The human remains the main brain: The shared study is not a coordinator or main brain. It is a shared context layer. The real main brain is still the human, plus a clear set of boundaries.
  • Boundaries matter more than connections: Without rules, a shared brain becomes a trash bin. The shared study should only accept team facts and long-term decisions. No customer-sensitive data, no raw chat transcripts, no passwords or credentials. Write access goes to only a few agents; the rest stay read-only.
  • Control plane separated from knowledge: The control plane tracks health and resource usage. The knowledge layer stores actual facts. When the shared study goes down, agents keep working; they lose team memory but do not die with it. The shared brain is a capability, not oxygen.

The four questions

Before adding agents, answer these:

  1. What belongs to a single agent and should never leave its room?
  2. What counts as a team fact every agent should be able to look up?
  3. Who has the right to write into the shared study?
  4. What can never enter the shared study, ever?

Seven-step implementation order

  1. Clean one agent's local memory. Trim its private notebook to what it actually uses.
  2. Build a shared team brain next to it. gbrain works well here.
  3. Wire the agent into the shared brain through MCP. Open search and read first. No write access yet.
  4. Let one or two clearly-owner agents write durable decisions into the shared brain.
  5. Add automated sync sources, but only sync curated content. Do not pipe in the whole inbox or Slack.
  6. Expand access by role. New agents default to read-only.
  7. Keep the control plane separate from the knowledge itself.

Evidence across sources

Source Key Claim Relevance
Vox — The First Step in Building My AI Native Team 先建共享记忆层再加 Agent;共享大脑是精选的团队事实层,边界比连接更重要 原始概念来源(个人/小团队尺度)
Every — Why We Replaced Personal Agents With One for the Whole Company 个人 Agent 舰队死于运维与知识孤岛;改为一个全公司共享 Agent 后,边界问题具体化为对话、连接、权限三条线的代码级规则 公司尺度的独立印证;把"边界"从记忆写入权扩展到凭证槽位与权限代码层

两个来源的互补:Vox 的框架回答"共享什么、谁能写",Every 的复盘回答"共享之后,行为层会坏在哪里"。Every 的"规则写在代码里、位于模型之下"可以读作本页"boundaries matter more than connections"的工程化版本。

Open questions

  • How does the shared brain pattern interact with the Memory Backend vs Context Substrate distinction? A shared brain could be implemented as either.
  • What happens when two agents disagree on a team fact? Is there a conflict resolution mechanism?
  • How does this scale beyond two agent platforms to tens or hundreds of agents?
  • Every 的"单共享 Agent"与本页的"多 Agent + 共享大脑"是同一原则的两种拓扑;什么条件决定选哪一种?

Prompts for witness

  • In your current setup, which decisions are worth writing into a shared brain, and which are just noise?
  • If you had to draw a hard boundary between what stays in a single skill/agent and what becomes team knowledge, where would you draw it?
  • 你的个人 Agent 工作流里,有哪些"只有你受益"的积累其实应该是团队级的共享事实?

2026-10-31 — Garry Tan:GBrain Cloud 作为跨 harness 共享上下文层

来源:AI Builders Digest 2026-10-31。

Y Combinator CEO Garry Tan 称 GBrain Cloud 让多个 agent 更方便地共用 GBrain。他点名 Grok、Muse、Dots 作个人用途,OpenClaw、Hermes、Claude Code、Codex、Capy 作更强或偏代码的 harness,把 GBrain 定位为跨 harness 的共享上下文层。

这是本页原则的第三次独立印证,且来自产品化一侧:前两个来源分别是个人/小团队尺度的设计原则(Vox)和公司尺度的复盘(Every),Garry Tan 的信号说明「共享大脑」正在从实践原则变成可售卖的基础设施层。他列出的 harness 名单也提示共享层要假设异构 harness 共存,而不是绑定单一平台。

Counterpoints & Gaps

  • 这是单条 X 动态,没有 GBrain Cloud 的访问控制、写入权限或冲突解决细节;产品是否能守住「boundaries matter more than connections」仍需独立验证。

Pending Jean review (2026-10-31)

  • harness-engineering/shared-brain-first — ai_draft; Garry Tan 信号可补入「Evidence across sources」表,建议在下次重写时合并进正文而非继续以追加段形式增长。

Sources

Synthesized from 4 sources
  • Vox — The First Step in Building My AI Native TeamSupporting source listed by this page.Whole pagemediumbody
  • Every — Why We Replaced Personal Agents With One for the Whole Company (2026-10-20)Supporting source listed by this page.Whole pagemediumbody
  • AI Builders Digest 2026-10-31 — Garry Tan 论 GBrain CloudSupporting source listed by this page.Whole pagemediumbody
  • 2026-10-31Supporting source listed by this page.Whole pagemediumabsorb log

Evolution

1 event
  1. absorbed

    Derived from source material

    This page is currently synthesized from 4 sources.

    From Vox — The First Step in Building My AI Native Team, Every — Why We Replaced Personal Agents With One for the Whole Company (2026-10-20), AI Builders Digest 2026-10-31 — Garry Tan 论 GBrain Cloud, 2026-10-31To Shared Brain First, Boundaries Second
    Sources: raw/social-triage/2026-05-13 The First Step in Building My AI Native Team.md · raw/newsletters/Every/2026-10-22 Why We Replaced Personal Agents With One for the Whole Company.md · raw/briefing/AI Builders Digest/2026-10-31.md · /Users/jean/Documents/Obsidian Vault/raw/briefing/AI Builders Digest/2026-10-31.md

Linked from