菜单
Mode B parallelism. Manually invoke Claude Code Agent Teams (multiple full Claude Code instances coordinating via shared task list + worktree isolation + Agent View). The multispec engine auto-selects this mode when the work demands it; this skill is for forcing it.
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
| name | agentteams |
| description | Mode B parallelism. Manually invoke Claude Code Agent Teams (multiple full Claude Code instances coordinating via shared task list + worktree isolation + Agent View). The multispec engine auto-selects this mode when the work demands it; this skill is for forcing it. |
Multiple full Claude Code instances coordinating via a shared task list with worktree isolation. Different from Mode A (SDK subagents in one query()) — these are separate processes, separate sessions, separate worktrees.
/agent-teams shipped experimental).CLAUDE_CODE_EXPERIMENTAL_AGENT_TEAMS=true (set automatically by .claude/settings.json).The multispec engine picks Mode B when ANY of:
5 sub-Specs
Override with cmax run "..." --mode teams or cmax run "..." --mode solo.
.claudemax/state/agent-teams-<ts>/shared-task-list.md. Every teammate has read/write..claude/worktrees/<session-id>/ automatically on first write.cd && claude --agent-view shows live state of all running sessions.16 agents wrote a Rust-based C compiler from scratch (100K LOC, ~2000 sessions, $20K credit) using this primitive. See https://www.anthropic.com/engineering/building-c-compiler.
Ask. Achieve. The canonical user-facing entry to claudemax — describe your goal, the SOTA-2026 pipeline (deepresearch → multispec → parallel /goal → verify) handles the rest. Same engine as /cmax with a friendlier verb. CLI equivalent is `cmax ask "<goal>"`.
The default FAT umbrella. Auto-runs deepresearch + multispec decomposition + parallel /goal + verify. Opus for planning/judgment/verification; Sonnet for sub-Spec execution (escalates per router). Multispec engine selects Mode A (SDK subagents) or Mode B (Claude Code Agent Teams) automatically based on work size. Your default daily-driver.
Max-effectiveness umbrella. Opus for EVERYTHING — planning, decomposition, sub-Spec /goal execution, verification. Use for novel domains, security/auth/payments, or when you want the highest ceiling regardless of cost.
Show the model-routing decision for a task (Fable/Opus/Sonnet/Haiku) with reasoning. Use before launching a packet to sanity-check the tier and to override when you have stronger judgment than the heuristic.
Reviews a proposed diff for correctness, security, style, taste alignment. Distinct from /audit (which reviews existing code, not diffs) and /verify (which checks SPEC completion).
Spec quality gate. Every completion condition must have a mechanically-checkable verifyHint (file path, command, test name, behavior). Blocks /goal handoff if the spec wouldn't be verifiable.