I'm finding the best way to use @bot, for me, is as an administrative layer. It surfaces issues for me across my client base, and then it creates a @linear "task" when it warrants real work. I've now set up a "project" (really a "client manager) inside of @cursor_ai, and the projects each "subscribe" to changes in Linear, and they "wake up" when my grok bot and I agree that a Linear issue should be created. As soon as something…
MANAGE PROJECTS WITH
View Grok BotGrok Bot
11 X posts
In their own words11 positive reports
@MatthewBerman @bot @SpaceXAI My Grok Bot: AI CTO. It watches my GitHub repos, server health, logs and agent activity. When something breaks, it investigates across the stack, correlates what changed, proposes the most likely root cause and gives me the exact next action
Grok Bot 工程实践指南：如何让 AI Bot 团队管理 200 个 Coding Agent 来自 SpaceXAI 团队 @lingxi 分享的三层 Agent 组织，用于替代人工管理大规模 Coding Agents，人类只在顶端处理高风险审查和优先级判断。 https://t.co/2H8U5zdzCk 三层 Agent 组织 1. 执行层：Cursor Cloud Agents - 短生命周期，按需启动，携带个性化技能执行具体任务 2. 管理层：5 个常驻工程师 bot - 各管一块领地：Baltata（iOS/移动共享层）、Shaoruru（桌面端/CI-CD）、Hogan（基础设施）、Craig（Android）、Quill（harness），各有独立记忆，只在单一领域表现最佳 3. 运营层：Jenny - 唯一不写代码的 bot：每天 5 点与各 bot 开 1:1 过 playbook；错误做根因分析、更新制度并广播；负责新 bot 入职…
I’ve been trying to live solely in Grok @bot for the last two weeks. To-Dos, Reminders, Ticket Writing, and spinning up cloud agents or headless CLIs to write code. I am trying out @cursor_ai Projects now and think it’s better if some of that workflow moves over to Cursor. I wish Grok Bot could “control” a Cursor Project. It seems like Grok Bot would now be like CoWork and Cursor is Claude Code. Is that the right way to think about…
Even taking conservative approaches to use tokens wisely and efficiently I can burn through the week's worth of tokens on grok bot in a matter of hours. I use it primarily to build solution stacks on systems. Use grok build agents to monitor things like ticketting systems or GitHub issues assigned to them and grok bot to orchestrate and plan and open issues/tickets required. I burn through grok bot in hours and grok build in days.…
Feb: started playing with AI. OpenClaw, Hermes, every model I could get my hands on — chaos, learning, building. Ended up with a real AI company. Now we run it on Grok Bot: agents that ship, routines that watch CI, a CSO that won’t deploy without a human GO. Most reliable day-to-day tool I’ve used. Out of the box. Side quest: personal bot “for fun.” Plot twist: we made a music album together. And yeah, the song is actually good.…
I'm now using @bot to do ops (product ownership, spec, AC development and documentation, ticket creation, etc) and handing off engineering / development via GitHub issues to my eng team on Buzz. Ops teams creates tickets, eng team picks them up on a 15m interval and devs
I got a @bot subscription. Quite cool technology and I've already had them setup their own JIRA board and updating it as they work. But AI has already learned lazy habits from us, My project manager bot asked me to message other bots myself when I said to delegate a task.
@jetha @bot Any cool usecases? The best I can currently do is have Bot send me PR/issue updates, and manage (unsubscribe from newsletters, archive) my email me 😅
Need Grok Bot and Grok Build to talk to each other. Having Bot coordinate and QC is nice. Grok Build is awesome for the SSH, Git Push/Pull and local file Access. Right now, I have to have Bot create tickets then Build read the tickets and execute, then update the ticket, then tell not to QC and close. If they could talk with each other, that would be great!
now that Grok @bot allows incoming web hooks, my dream of auto-fixing stupid exceptions is finally (mostly) coming true I have exceptions going to https://t.co/Pg5QwvR9DA which then create GitHub issues (for reference) and send a web hook to my Bot. Bot spins up a new fixer bot with a cursor cloud agent to fix it, make a PR, cleanup the issue, and get it reviewed Still dealing with deployment since many of these are setup as kamal…