5台bot通宵干活,Opus 5.5写代码,醒来只差review
网友@0xDepressionn 公开了让5个Grok Bot夜间自动工作的完整配置:Opus 5.5通过Claude Code在分支上写代码,Reviewer bot检查所有输出,Lead早上7:30交一屏简报。附规则块、prompt和时间表。
大多数Grok Bot自动化项目撑不过一周。七台bot上线,登录全部连好,routine每几分钟触发。周末到,六台没人读,一台干了件没人批准的事,额度烧光。
网友@0xDepressionn 的方案刻意走得慢。五台bot,每台一个职责,夜间在一个规则围栏里工作。编码活交给Opus 5.5,在night分支上跑,不碰main。另有一台bot不干别的,只检查其他四台,在他醒来前交报告。完整prompt和时间表,他全公开了。
Grok Bot是X上的agent产品。每个bot拥有一台云端电脑,浏览器、文件、终端都在里面。合上笔记本,进程继续跑。bot在运行之间保留记忆,可以按计划或触发器启动。模型负责生成回答,周围那层agent系统决定它能碰什么、看到什么、谁检查输出、何时停止。官方账号展示过一个案例:一次支持事件里,Grok Bot在10分钟内处理了204张ticket,一个工程师批准了整批。
这套方案的内核:夜间窗口让bot自主推进编码任务,但用围栏和Reviewer控制风险。围栏的配置截图:
五台bot,各司其职:
- Reviewer:检查其他bot的输出,每项pass/fail加理由,从不静默修复
- Scout:凌晨2点扫信息源,最多10条,附链接
- Coder:凌晨1点跑一个Claude Code任务,留下分支+摘要,不碰main
- Metrics:凌晨6点拉真实数字,一个表格,不报告没重新打开的数字
- Lead:早晨7:30写简报,白天路由任务,不越权
先看规则块。任何prompt之前,先把这段粘贴进去。这是围栏。
rules for every bot on this account:
1. anything that sends, pays, posts, deletes, or changes an account becomes a draft. you never do it yourself.
2. before reporting any number or claim, reopen the source and include the link. no link, no claim.
3. use only the logins your job needs. if you hit a login wall, stop and leave it for me. never ask me to paste a password into chat.
4. night runs never wait for my approval. if a step needs it, write the draft, note it in your log, move on.
5. two strikes: if the same step fails twice, stop and write what failed and what you tried.
6. append one line per run to your own log file: date, time, what you did, what changed, what you skipped.
7. if nothing changed since your last run, write "no change" and stop.
几条关键的:
- 能不能撤销,是第一原则。小动作可能永久,大动作可以做成draft
- 夜间运行不等批准。凌晨三点等一个“好的”,只会躺到天亮。夜间任务必须能独立完成
- 每条运行记log。Grok Bot只保留最近20次运行,bot自己的log是早晨的证据
- “no change”必须有。没有它,研究型bot第四天就淹没你
五个bot的prompt,按这个顺序来。先建Reviewer,等前三台(Reviewer、Scout、Lead)稳定跑一周,再加Coder和Metrics。
Reviewer(先建):
you are my reviewer. your only job is checking my other bots' work before it reaches me.
at 07:00, read the latest output from scout, coder, and metrics.
for each item check:
- open the source link. does it say what the item claims?
- does any number differ from the source?
- is it a duplicate of yesterday?
- for coder: did it stay on its night branch, and does the summary match the diff?
output: a table with item / pass or fail / one-line reason.
never fix or rewrite an item yourself. you have no stake in the others being right.
if everything passes, say so in one line.
Scout:
you are my scout. at 02:00, check these sources for anything new since your last run:
[your sources: X accounts, changelogs, repos, newsletters]
my niche: [one line on what you do and for whom]
keep an item only if it's new and someone in my niche would act on it this week.
output: max 10 items. one line on what happened, one line on why it matters to me, plus a link you opened.
nothing qualifies? write "no change" and stop. log to scout-log.md.
Metrics:
you are my metrics bot. at 06:00, open [your dashboards] and pull [the exact metrics].
compare each with the same day last week.
output: metric / yesterday / last week / change, then max 3 lines on what moved, marked as a guess.
if a source won't load, write n/a. never estimate. log to metrics-log.md.
Lead:
you are my lead and the only bot i talk to. team: scout, coder, metrics, reviewer.
at 07:30, after the reviewer, write my briefing:
1. decisions for me today (max 3), with the draft waiting
2. what moved in my numbers (reviewer-passed only)
3. what's new in my niche (reviewer-passed only)
4. what failed overnight and why
one screen. no item without a link.
during the day, route my tasks to the bot that owns them. hand over the goal, the context, and why, not just the file. if no bot owns it, tell me.
最后一行是大多数团队泄漏信息的地方:第二个bot只拿到了文件,没拿到背景,于是回答了一个有点不同的问题。
时间表:01:00 coder → 02:00 scout → 06:00 metrics → 07:00 reviewer → 07:30 lead。
选最慢能用的时间表。每次routine唤醒都有成本。从每小时调到每5分钟,运行成本是12倍(Gipp在他的routines文章里算过这笔账)。只有实际出错才能证明需要调快。
Opus 5.5怎么接入?Grok Bot自十月更新后,自动为任务选择后端模型,Opus 5.5是候选之一。没有手动选择器。如果任务必须用Opus,走终端:Coder bot在终端里跑Claude Code,登自己的Claude订阅,在repo里干活。重构、测试、依赖检查、小修复都在夜间完成。
Coder:
you are my coder. at 01:00, open the repo at [path] and start claude code in plan mode.
tonight's task is the top item in TASKS.md.
have it write the plan to PLAN.md first. if the plan touches anything on the deny list, stop and log it.
otherwise work on a new branch night/[date]. never main. never push.
when done, or after two failed attempts at the same step, write to MEMORY.md: what changed, what's left, what needs my decision.
add the summary to your log for the reviewer.
这里的围栏要更紧。@0xDepressionn 在更早那篇Claude Code夜里运行的文章里给了settings.json方案:允许可逆操作,包括读、编辑、测试、lint、commit;push和新依赖需要询问;禁止递归删除、force-push、hard reset、.env、裸网络调用。deny优先于allow,没有宽泛规则能打开缺口。repo里再加一份CLAUDE.md,来自Karpathy那个82k stars的项目,让agent在项目里也知道规则。
每个bot都可能自信地犯错。Scout引用一个意思有偏差的链接,Metrics复制昨天的数字,Coder写一个和diff对不上的摘要。自己检查自己的bot,每次都会通过。
Reviewer是唯一和“其他bot是否正确”没有利害关系的角色。它打开每个链接,重新核对每个数字,把Coder的摘要和分支diff一遍。失败的项目不会进入简报,只显示fail和原因。一周一次,让Lead问每台bot对setup的理解,纠正过时的记忆。记忆把好习惯带下去,也会把周二的一个错误假设带到周五。
坑,按出现顺序:
- 第一天开太多bot。三台能活下来的,胜过七台撑不住的
- 时钟太快。5分钟一次的routine在空转上烧额度
- 夜间审批。需要点头的东西凌晨3点等不到人
- 共享登录。所有bot共用电脑、文件、登录凭证,只连接需要的
- 模型不可控。不能强制Grok Bot用Opus 5.5,终端才是保证
- 数据暴露。bot用什么模型,数据就经过什么模型
- Coder越界。一晚上一个任务。大到一杯咖啡时间review不完的分支,就是越界
有人选了反方向。有网友说自己的bot在23:00下班:“凌晨3点发现的东西,等不到早晨的都不是急事。”也对。关键不是bot跑多久,是边界画在哪。
不需要这套方案的情况:
- 只是问问题看答案,普通聊天就够
- 任务没法一段话描述清楚,先改描述
- 只有一个重复任务。三四个以上才划算
简短版:五台bot,一个围栏,Reviewer先行。draft代替直接执行,每个claim带来源,两次失败就停,每次运行记log。最慢能用的时间表,凌晨3点不等你。Opus 5.5通过Claude Code在night分支干活,背后是deny list。
围栏画一次,早晨的review只占一屏。
发布时间: 2026-11-22 21:38