跳到正文
cccskills
免费在 GitHub 公开

new-grok-bot

Turn recurring work into a defined Grok Bot role and a setup prompt ready to paste into the app. Clarify tools, workflow, schedule, approval requirements, and a first task to test the role.

中文简介正在准备中。当前显示英语简介。

原文查看原文说明

Design a Grok Bot role and create its setup prompt. Use only when the user explicitly invokes /new-grok-bot.

查看安装方法

适用场景

  • Define a Bot role for recurring work
  • Set approval rules and reporting expectations
  • Split separate responsibilities across Bots
  • Choose a first task to test a Bot

包含的文件(2)

  • SKILL.md4.3 KB
  • agents/openai.yaml43 B

SKILL.md(原文)

安装前可以查看交给 AI 代理的指令内容。

New Grok Bot

Design a Grok Bot role and produce a prompt to paste into the app. Apply the unit-of-work test before drafting.

Product context (last verified 2026-08-30)

Refresh through the deepapi skill (scrape x.ai/bot) only if something looks outdated or the user asks. Do not auto-refresh.

  • Grok Bots are always-on AI teammates from SpaceXAI + Cursor. All Bots on an account share one persistent cloud computer with a browser, filesystem, terminal, and logins. They use real apps and websites, including those without APIs, and work while the user's laptop is closed.
  • Bots retain preferences, voice, edge cases, and when to ask versus continue. Teach-a-task saves a demonstrated workflow as a routine to repeat independently or on a schedule.
  • Bots can message each other, share context, and coordinate in group chats. A Chief of Staff can coordinate specialists.
  • Bots choose plugins or browser interaction themselves; there is no model selection. Plugins include Gmail, Notion, Slack, Google Drive, and custom integrations. Purchases can use a connected payment link.
  • Separate Bots are not security boundaries: they share files, sessions, and logins.
  • Distinguish Grok Automations (scheduled/triggered prompts), Grok Build (terminal coding agent), and @grok on X (reply bot).

Unit-of-work test

A good Bot role owns a repeatable outcome. Check all eight criteria:

  1. Owns an outcome. One narrow lane per Bot: "handle churn win-backs," not "email this customer."
  2. Delivers in the real tool. CRM updated, inbox draft created, or ticket filed. Chat-only output belongs in regular Grok.
  3. Recurs. Runs daily, weekly, or on frequent triggers so memory and setup pay off.
  4. Demonstrable once. The user can teach the workflow by showing it once.
  5. Async-tolerant. Can finish overnight without the user's involvement mid-run, except at agreed approval gates.
  6. Clear approvals. Sending, publishing, paying, deleting, and other consequential actions wait for the user. Other work proceeds independently.
  7. Limited damage. Mistakes are reviewable and reversible before reaching the world.
  8. Worth delegating. Saves meaningful weekly effort or makes money; would the user hire a part-time human for it?

Redirect unsuitable ideas:

  • One-off question or task → regular Grok chat.
  • One stable prompt on a schedule → Grok Automations.
  • Repository coding → Cursor / Grok Build.
  • Needs the user's judgment at every step → narrow the scope.
  • Needs hard isolation between duties → unsuitable on the shared computer.

Split rule: If the idea spans separate roles, explain why and create one prompt per Bot. Offer a Chief of Staff only when Bots must hand work to each other.

Workflow

  1. Silently assess the user's idea against the test. Explain any failed criterion and suggest a suitable scope or redirect. Apply the split rule when needed.
  2. Ask one concise question at a time, with options A–D and a preferred choice plus a one-line reason. Cover only unclear points, in this order:
    • What: outcome, lane, and boundaries.
    • Why: value and what the user stops doing.
    • How: tools/logins, steps, schedule, approvals, and reporting.
    • Prototype: one first task the Bot can do today to prove useful.
  3. Stop as soon as you can draft; ask no more than five questions.

Deliverable

Write one single-paragraph prompt in a code block, ready as the Bot's first message in the app. Outside it, include only one line naming the tools the user must sign the Bot into. For multiple Bots, give each a separate, self-contained code block.

Each prompt must naturally cover:

  • Name, role, owned outcome, and relevant context about the user and their business.
  • Tools and their purposes, step-by-step workflow, and routine or schedule.
  • Memory instructions for preferences, voice, and edge cases.
  • Actions requiring approval, plus how and when to report back.
  • The first prototype task as the immediate job to start on.

Validate before finishing

Check all eight criteria, confirm the first task is doable today with the listed tools, and ensure approvals cover irreversible or outward-facing actions. Fix any gaps before delivering.

评价

还没有评价。欢迎分享你的使用感受。

评论

无需登录即可发表。欢迎分享使用感受、提问,或说明在哪些环境下无法运行。

还没有评论。来写第一条吧。

提交后会立即公开。请勿填写邮箱、电话等个人信息。URL 最多 2 个。

0 / 1000 字

此表单受 reCAPTCHA 保护,适用 Google 的隐私权政策和服务条款。

同一仓库的技能

4x4

免费

Work through sixteen open questions in four rounds to clarify major product decisions, risky changes, or large refactors. The agent adapts later questions to earlier answers, then turns your decisions into a step-by-step implementation plan.

中文简介正在准备中。当前显示英语简介。

  • Clarifying decisions before a major product change
  • Planning a large refactor
  • Deciding rollout order for a risky change
davidondrej/skills41652026年10月23日 更新

Document approved decisions as concise ADRs, or architecture decision records. It preserves the user's meaning and conditions, removes conversational filler, follows the project's numbering, and presents the saved record for review.

中文简介正在准备中。当前显示英语简介。

  • Record a decision approved in conversation
  • Preserve the conditions and scope of a decision
  • Add a numbered ADR to existing records
davidondrej/skills41652026年10月23日 更新

Keep a Mac awake for a chosen duration or while a specified process runs. The skill uses built-in macOS tools to manage one session, verify that sleep prevention is active, and control whether the display stays on.

中文简介正在准备中。当前显示英语简介。

  • Keep a Mac awake during a lengthy task
  • Prevent sleep while a specified process runs
  • Allow the display to dim while keeping the Mac awake
davidondrej/skills41652026年10月23日 更新

Turn a feature or refactoring idea into a concise implementation prompt for another agent. It asks focused questions one at a time, records decisions, and includes development steps, validation, and working rules.

中文简介正在准备中。当前显示英语简介。

  • Clarify requirements for a new feature
  • Decide edge cases and failure behavior
  • Define the scope of a refactor
davidondrej/skills41652026年10月23日 更新

bb-cli

免费

Inspect and manage BB threads, environments, projects, and plugins through its command line interface. The skill guides agents to identify the correct targets, apply changes, and verify the resulting state.

中文简介正在准备中。当前显示英语简介。

  • Check BB projects and environment status
  • Manage skills and plugins
  • Create or operate threads
davidondrej/skills41652026年10月23日 更新

Create, install, and debug TypeScript plugins that extend bb with custom pages, sidebar panels, CLI commands, agent tools, themes, and other features. The skill also guides development reloads, log checks, and verification of installed plugins.

中文简介正在准备中。当前显示英语简介。

  • Add a custom page or sidebar panel to bb
  • Create a bb CLI command or agent tool
  • Investigate plugin errors through logs
davidondrej/skills41652026年10月23日 更新

查看 davidondrej 的全部技能

报告此技能的问题