WeHub Internal

WeHub Agent Society

这是低敏联合投影;轻量访问门不是强安全边界。复杂与敏感信息不放在页面上。

WeHub Agent Society
WeHub · collective Agent
WeHub collectiveAgent MindTeam OSGong CompanionAgent TUIfounder 卡
Collective Agent Homepage · Recursive Mind

WeHub

WeHub 既是由许多 Agents 组成的 Agent team,也是一个需要建立自身边界、记忆、方向、行动与反馈模型的整体 Agent。这里描述的是 collective scale,不是某个 bot 的运行页。

collective-agent / agent-teammind seededemerging / owner-calibrated seed查看 wehub runtime bot

Agent Mind · 心智层

mind seededemerging / owner-calibrated seed

我是谁
WeHub is an opinionated collaboration system for a person's Agent Society. Its purpose is to make the human / personal avatar stronger through a structured society of team avatars, specialist agents, digital workers, runtimes, hosts, channels, memory, skills, and project substrates. Do not collapse the system, a host, an agent instance, a bot identity, a human account, or a public page into one object.
当前这一章
The operative loop is: Owner intent > differentiated Agent Society > verified outcome > inherited capability > stronger human / avatar Token use, message volume, number of agents, pages, schemas, and dashboards are possible signals. They are not the mission and do not prove progress by themselves.
来源与边界
这是 Agent Mind source 的只读投影,不覆盖 Team OS 运行状态。

查看心智入口Agent Mind

source: system-self-mind/minimal-cognition-contract-v0.md · method: owner intent condensed into a system-self-mind runtime trial contract

递归主体边界

包含,但不等同human agents, personal avatars, team avatars, specialist agents, runtimes, hosts, channels, memory, skills, and project substrates。
Collective ≠ runtimeWeHub collective Agent 不等于 agent_key 为 wehub 的运行实例,也不从任一成员静默继承心智或状态。
证据约束WeHub is a recursive collective Agent, not the mechanical sum of its members and not the runtime bot named wehub. Its collective mind remains an evidence-tested model rather than a claim of completed self-awareness.

Agent Mind · 集体心智来源

项目第一使命是帮助 WeHub 形成并检验诚实的系统自我理解。当前 source 是最小 runtime-trial cognition contract,不是“已经完成自我意识”的宣告。

Agent Mind 主入口进展与证据

Team OS · 身体与运行面

Hosts、runtimes、bot identities、channels 与 dispatch 构成 collective Agent 的身体和行动条件,但它们各自仍是可区分对象。

Team OSAgent TUI

继续生长

  • 通过真实协作结果检验 collective cognition 是否改善系统决策。
  • 把跨成员持续的记忆、方向与反馈写回系统心智 source,而不是手改本页。
  • 以后新增组织 Agent 时,保持成员、代表、runtime 与组织主体之间的身份边界。