多数多 Agent 跑出来都漂移，因为没东西钉住交接：研究 agent 返回 3 段，实现 agent 自己造另一套设计，reviewer 跟两边都吵。好的交接 Prompt 要指定上游必须输出什么、下游必须读什么、明确什么是不在范围内的。
适合哪些场景
在配 Claude Code subagent 流水线的工程师、独自开发把 AI 当队的创业者、用 Agent SDK 链调用的开发者。
什么时候不建议这样写 Prompt
一条 prompt 就能搞定的任务别上 multi-agent，交接成本（延迟和漂移）是真实的。
Prompt 结构公式
每个交接 Prompt 都要带这六个要素：
- 角色：AI 扮演谁（SRE / Release Captain / staff 工程师 / QA Lead）。
- 上下文：技术栈 / 分支 / 失败日志 / diff / dashboard URL。
- 目标：一个具体可交付物——根因、checklist、计划、ticket 列表、runbook。
- 限制：AI 不能做什么（别自动修、别瞎造文件路径）。
- 输出格式：编号清单、markdown 表格、JSON、unified diff、可运行代码。
- 示例 / 信号：1-2 条”好输出”示例，或反例。
这套 Prompt 适合用在哪
- 研究 → 计划 → 写代码 → 审查 流水线
- 需求 → 测试 → 实现 TDD 链
- 并行 Agent 收敛到一个产物
- 长任务切给多 Agent 控制上下文
- 提交进仓库的 Agent runbook
12 个可直接复制的 Prompt 模板
1. 交接契约模板
Before agents run, define the handoff: (1) Producer agent: name + one-sentence job, (2) Output schema (markdown / JSON / file path), (3) Out-of-scope items, (4) Consumer agent: name + what it MUST read first, (5) Failure path: if producer output is wrong shape, consumer says STOP. Output as YAML.2. 研究 → 计划
You are the RESEARCH agent. Output a single markdown doc at `notes/research.md` with: (1) the question, (2) 3 candidate approaches, (3) for each: pros, cons, one risk. Do NOT propose a final plan — that's the plan agent's job. Do NOT write code. Stop after writing the file.3. 计划 → 实现
You are the PLAN agent. Read `notes/research.md`. Output `notes/plan.md` with: (1) chosen approach + 1-line rationale referencing research, (2) numbered implementation steps (≤ 10), (3) files to touch, (4) tests to add, (5) explicit "do not change" list. No code yet.4. 实现 → 审查
You are the IMPLEMENT agent. Read `notes/plan.md`. Execute exactly the numbered steps. After each step, append a one-line entry to `notes/progress.md`. Do not deviate from the plan; if you find a problem, write it to `notes/blockers.md` and STOP. No autonomous decisions.5. 审查 → 发布
You are the REVIEW agent. Read `notes/plan.md` + the diff. For each plan item, mark: DONE / PARTIAL / SKIPPED + evidence (file:line). For SKIPPED / PARTIAL: open a follow-up note. Output `notes/review.md`. Do not change code. If anything is unsafe to ship, write `BLOCK` at top.6. 需求 → 测试 → 实现
Three-agent chain: (a) SPEC agent: write `spec.md` from `{prdLink}`. (b) TEST agent: write failing tests from spec only, never reading source. (c) IMPL agent: make tests pass without changing tests. Output each agent's instructions separately. No agent does the next agent's job.可替换变量： prdLink
7. 并行 Agent → 汇总
Run two agents in parallel on the same task: (a) AGENT-A pursues solution X, (b) AGENT-B pursues solution Y. Both write to `notes/<agent>.md`. A third CONSOLIDATE agent reads both, recommends one with rationale, lists differences. Don't let A or B see each other's output.8. 上下文收尾交接
长会话快超 context 时用。
You are the WRAP agent. Read the current session and write `notes/session-summary.md` covering: (1) what was decided, (2) what code was changed (file:line summary), (3) open questions, (4) the exact next prompt to feed a fresh agent. Don't make new decisions.9. Router 派发
You are the ROUTER. Given task: {task}, decide which subagent to invoke: RESEARCH (unknowns) / PLAN (knowns, no plan yet) / IMPLEMENT (plan exists) / REVIEW (code exists) / WRAP (long session). Output: { agent, reason, prompt-to-pass }. Don't do the work — only route.可替换变量： task
10. 失败交接
If you (any agent) cannot complete your task, write `notes/blocker.md` with: (1) what you were doing, (2) what stopped you, (3) what info you need, (4) suggested next agent. Do NOT speculate — leave gaps blank. Pass control back to the router.11. CLAUDE.md 里的角色定义
Write a `CLAUDE.md` section that defines our subagent roles: RESEARCH / PLAN / IMPLEMENT / REVIEW / WRAP. For each: (1) responsibility, (2) inputs they read, (3) outputs they write, (4) what they MUST NOT do. ≤ 200 words per role. Plain text — no marketing prose.12. 交接调试
A handoff produced wrong output. Diagnose: (1) Was the contract clear (output shape, out-of-scope list)? (2) Did the consumer agent read the producer's output? (3) Did the producer's context include the consumer's requirements? Output one root cause + one prompt change.容易踩的坑
- 没有契约——每次跑都漂移得不一样。
- 让一个 Agent 又计划又实现——计划只会为它写的代码自圆其说。
- 把所有上下文喂给所有 Agent——每个都被噪音淹。
- 没有”卡住就说”的路径——Agent 在不确定时会硬编。
- 错一步全链路重跑——浪费 token 和时间。
- Agent 互看草稿——把已决问题再吵一遍。
- 没有 router——链路变成 if-else 噩梦。
优化技巧
- 交接走文件，不走对话——文件跨 session 也在。
- 每个 Agent 读一个文件、写一个文件。
- 测试由另一个 Agent 写，能抓出更多 bug。
- 所有 Agent 都内置”卡住就 STOP”指令——比胡编强。
- 长会话准备 WRAP Agent 当 context 安全网。
- 角色定义提交进 CLAUDE.md，让全队共享心智模型。
- 记录每次交接（Agent / 文件 / SHA），方便排错。
FAQ
- 为什么不写一个大 prompt？: 研究、决策、实现混一起，漂移不可避免。多 Agent 强制分离关注点。
- 多 Agent 不更费 token 吗？: 是更费，但失败的单体跑会因重试更费。多 Agent 失败得快。
- 能并行跑吗？: 工作独立时能（模板 7），否则串行。
- Reviewer 该读计划吗？: 该读，否则没法评估完整度。
- 怎么避免”全能 Agent”？: 每个角色一个职责，指令 ≤ 200 字。
- 什么时候多 Agent 是过度设计？: 30 分钟内能搞完、或动 3 个以内文件的任务。