agent-team-work-zone

Team lead invokes after session restart: reads TEAMMATE_INFO.json to rebuild the team by using the Agent tool to spawn fresh sessions for each active teammate (name must be <slug>-<role>; permission mode inherits the lead, not set per-teammate at spawn), guiding each to read its workstation's working-context.md (Part A snapshot + Part B work journal) to self-recover state. No-arg invocation wakes only active/idle and **skips benched** (temporarily offline); `/reactivate-team <name>` wakes back one specified (usually benched) teammate. This is the ONLY way to "recover a team across sessions" — Claude Code does not auto-respawn teammates.

Structured flow for a team lead to assemble a Claude Code agent team. 6 phases: task decomposition → lineup proposal (selected from role_archetypes) → plan-mode gating → adversarial check → user confirmation → spawn each teammate via the Agent tool (name must be <slug>-<role>; permission mode inherits the lead, not set per-teammate at spawn) + save team_recipe. After Phase 5 user confirmation, Phase 6 spawns immediately with no additional confirmation. The agent can invoke autonomously; if currently a flat workstation, it will first guide through /promote-to-team.

Team lead 在 session 重启后调用，读 TEAMMATE_INFO.json 重建 team： 为每个 active teammate 用 Agent 工具 spawn 新 session（name 必须为 <slug>-<role>；权限模式继承 lead，不在 spawn 时单设）， 引导它读自己工位的 working-context.md（Part A 快照 + Part B 工作日志）恢复状态。 无参调用只唤醒 active/idle，**跳过 benched**（临时下线）；`/reactivate-team <name>` 单独唤回指定的（通常是 benched）teammate。这是"跨 session 团队恢复"的唯一方式—— Claude Code 不会自动 respawn teammate。

Team lead 组建 Claude Code agent team 的结构化流程。6 阶段：任务分解 → 阵容提案 (从 role_archetypes 选) → plan-mode gating → 对抗性检查 → 用户确认 → 用 Agent 工具（name 必须为 <slug>-<role>；权限模式继承 lead，不在 spawn 时单设）逐一 spawn 每个 teammate + 保存 team_recipe。Phase 5 用户确认后直接 spawn，不再额外确认。Agent 可自主调用；若当前是扁平工位，会先引导 /promote-to-team。

Add a teammate to an existing team: runs a simplified spawn-team flow (single person only). Team lead can invoke autonomously (after natural-language user consent). Valid only in team lead context.

Temporarily take a teammate offline (benched): have it write a final checkpoint, close its session to free an online slot, but **keep its full record + workstation + docs**, and set its status to benched in TEAMMATE_INFO.json. Unlike /remove-teammate (permanent offboard), benched means "will come back" — wake it later with `/reactivate-team <name>`. Team lead invokes autonomously (after user consent). Only valid in team lead context.

Teammate persists current work state to its workstation's working-context.md: Part A current-state snapshot (overwrite) + Part B work journal (append; recent conversation and verbatim key exchanges), and appends completed.md as needed. Triggered automatically by the TeammateIdle hook (working-context.md mtime gate + exit 2 reminder), or by explicit lead request, or manually. **Not /compact** — does not destroy current context. Rule 13 mandates that teammates call this skill regularly.

Onboard a new agent: first ask whether it's a flat workstation or team lead, then create the corresponding directory structure, generate the role definition file and task tracking files, and register in the member table. Use at the start of a new agent conversation.

Promote a flat workstation to team lead: rename the directory to <name>_team/, add roundtable/archive/team_recipes/teammates/ subdirectories, modify README to add rule 12 and lead-specific sections. The agent can invoke autonomously—when a flat agent foresees the task will become complex, proactively suggest to the user.

在现有 team 中增加一个 teammate：走简化版 spawn-team 流程（只处理单人）。 Team lead 可自主调用（用户自然语言同意后）。只在 team lead 上下文有效。

把某个 teammate 临时下线（benched）：让它做最终 checkpoint、关闭其 session 腾出在线名额， 但**保留全量档案 + 工位 + 文档**，并在 TEAMMATE_INFO.json 里把 status 置为 benched。 与 /remove-teammate（永久下岗）不同——benched 语义是"还会回来"，日后用 `/reactivate-team <name>` 单独唤回。Team lead 自主调用（用户同意后）。只在 team lead 上下文有效。

Teammate 把当前工作状态写到自己工位的 working-context.md：Part A 当前态快照（覆写） + Part B 工作日志（追加，含近期对话与关键往来原文），并按需追加 completed.md。 由 TeammateIdle hook（working-context.md mtime 闸门 + exit 2 提醒）自动触发， 或 lead 明确要求，或手动调用。**不是 /compact**——不破坏当前 context。 Rule 13 规定 teammate 有义务定期调用本 skill。

新 agent 入职: 先询问是扁平工位还是 team lead，然后创建对应目录结构、 生成角色定义文件和任务跟踪文件、在成员表中注册。在新 agent 对话开始时使用。

把扁平工位升级为 team lead：重命名目录为 <name>_team/、补齐 roundtable/archive/ team_recipes/teammates/ 子结构、改 README 加 rule 12 和 lead 专属章节。 agent 可自主调用——当扁平 agent 预见任务将变复杂时主动向用户建议。

Scan the top-level meeting_room (all agents) + your team's roundtable (team leads only) for completed (RESOLVED) files; by permission rules archive only files the current agent has authority to handle, moving them to the corresponding layer's archive. Strictly follow work rule #8 (including cc field). (From v0.3.0, prefer /check-inbox instead — see DEPRECATED NOTICE)

Check tasks addressed to you in the top-level meeting_room (all agents) + your team's roundtable (team leads only); process in chronological order: read content, update workstation files, execute tasks, update status. Also scan as issuer for your own completed docs and archive them (step 9). Strictly follow work rule #8 (including cc field semantics). Authoritative archival implementation; /archive-resolved is deprecated.

Team lead evaluates the current team's efficiency and composition: who is busy, who idle, whether a role is missing, whether someone can be let go. Produces a structured evaluation report to help the lead decide whether to /add-teammate or /remove-teammate. The agent can invoke autonomously (after natural-language user consent).

Task handoff: a single skill supporting both the giver (generate handoff document) and the receiver (read handoff document and absorb tasks) modes. Ask the user at the start which they are, then enter the corresponding flow. Handoff documents follow the top-level meeting_room frontmatter conventions, adding a new kind: HANDOFF type. Used in scenarios involving agent responsibility change or task migration after refactoring. Treat the task as a black box: this skill does not care how the receiver will complete the task.

Offboard a teammate: gracefully complete handoff, archive their output, remove from team. Team lead can invoke autonomously (after natural-language user consent). Valid only in team lead context.

Sync workspace changes and restore context. Scans project team member changes, new meeting room messages, and department structure updates; identifies content needing adjustment and generates an action list. Team leads additionally scan their department's roundtable. Also used for role recovery after context compaction.

扫描顶层 meeting_room（所有 agent）+ 自己 team 的 roundtable（仅 team lead）中 已完成 (RESOLVED) 的文件，按权限规则只归档当前 agent 有权处理的文件， 分别移动到对应层级的 archive。严格遵守工作守则第 8 条（含 cc 字段）。 （v0.3.0 起建议用 /check-inbox 代替，详见 DEPRECATED NOTICE）

检查顶层 meeting_room（所有 agent）+ 自己 team 的 roundtable（仅 team lead）中 指向自己的任务，按时间顺序处理：读取内容、更新工位文件、执行任务、更新状态。 同时作为 issuer 扫描自己发出的已完成文档并归档（步骤 9）。 严格遵守工作守则第 8 条（含 cc 字段语义）。归档权威实现；/archive-resolved 已 deprecated。

Team lead 评估当前 team 的效率和人员构成：谁在忙、谁闲、是否缺某种角色、是否有人可裁。 产出一份结构化评估报告，辅助 lead 决定是否 /add-teammate 或 /remove-teammate。 agent 可自主调用（用户自然语言同意后）。

任务交接：单个 skill 支持交出方（生成交接文档）和接收方（读取交接文档并吸收任务） 两种模式。开头询问使用者身份，然后进入对应流程。交接文档遵守顶层 meeting_room 的 frontmatter 约定，新增 kind: HANDOFF 类型。用于 agent 职责变更或重构后的任务迁移 场景。把任务当成黑箱：本 skill 不关心接收方将如何完成任务。

让某个 teammate 下岗：优雅完成交接、归档其产出、从 team 中移除。 Team lead 可自主调用（用户自然语言同意后）。只在 team lead 上下文有效。

同步工作区变更并恢复上下文。扫描项目组成员变动、会议室新消息、部门结构更新， 识别需要调整的内容并生成行动清单。Team lead 会额外扫自己部门的 roundtable。 也用于上下文压缩后的角色恢复。