Bot Generator Kit
Turn a job you do by hand into a safe, tested AI teammate in under an hour.
GetKits 제작
세 가지가 함께 돌아가는 모습
요청하면 Skill이 일을 처리하고, Bot이 Grok에서 루틴을 돌리고, Dot이 ChatGPT에서 계속 지켜봐요. 아래 각 단계는 이 Kit에 들어 있는 실제 전후 비교예요.
사용 전
Every Friday I match last week's card charges to receipts and chase missing ones, 50 min. Receipts in Gmail, card CSV in Drive. Write to my Expenses 2026 sheet. Never email vendors without me, never touch past months. Friday 4pm, Lisbon.
사용 후
Ship, 16/16. Expense Manager (Grok Bot), every Friday 4:00 PM Europe/Lisbon: matches charges to receipts in Expenses 2026, keyed by Transaction ID, and drafts one chase email per merchant. Nothing is sent. No CSV modified in the last 7 days means ROUTINE FAILED, never last week's file. Time saved: 40 min/week. Comes with the profile, 3 skills, 2 routines and a test run on a copy of the sheet.
차이를 직접 확인하세요
From a Friday scout report to a kept skill in 30 days
사용 전
Dot, Fri Oct 16, 2:00 PM: 4 proposal follow-ups written by hand this week, same 3 paragraphs, about 9 minutes each.
사용 후
Kept after 30 days, 25 min/week saved. The dot spotted it, the Bot scored it (Priority 20) and logged it in Notion, the Skill built proposal-follow-up and passed its trust audit 16/16, and the Monday 30-day check confirmed it. Drafts only: you still send every follow-up yourself.
One Friday: the dot finds it, the Bot scores it, the Skill builds it
사용 전
Fri Oct 16, 2:00 PM, from the dot's SCOUT REPORT: NEXT AUTOMATION Job: Proposal Follow-up: drafts the follow-up 3 days after you send a proposal Evidence: 4 follow-ups in Sent, Oct 12 to Oct 15, same 3 paragraphs Time saved: 28 min/week (9 - 2) x 4, estimate 3:00 PM, Bot Generator asks its three questions. You paste the card as answer 1, then: 2) nothing new 3) copied the 4 proposal dates into my pipeline sheet by hand, 5 min each
사용 후
BACKLOG UPDATE (Automation backlog) 1. Proposal Follow-up: drafts the follow-up 3 days after a proposal | 28 min/week | Skill | Low | Priority 20 | Interviewed 2. Pipeline Logger: copies proposal dates into the pipeline sheet | 12 min/week | Grok Bot | Medium | Priority 10 | Idea (under 15 min/week, parked) NEXT AUTOMATION Job: Proposal Follow-up: drafts the follow-up 3 days after you send a proposal Evidence: 4 in Sent, Oct 12 to Oct 15 (dot), confirmed by you Time saved: 28 min/week (9 - 2) x 4, estimate Form: Skill now; after 3 good runs, a Grok Bot routine can draft each one on day 3 Risk: Low (drafts only) First build step: paste the INTERVIEW below into the Bot Generator Skill Decision: build | skip | later INTERVIEW Job: Proposal Follow-up: drafts the follow-up email 3 days after you send a proposal Inputs: the proposal thread, pasted by you or found by subject [confirm: subject pattern, for example "Proposal:"] Tools: Gmail: read Good output: [confirm: paste your best follow-up of the four] Never: send; promise discounts, dates or scope How often: about 4 a week, on request, 9 minutes by hand Open: [confirm: subject pattern]; [confirm: best follow-up]
The monthly METRICS block, eight weeks in
사용 전
Automation backlog, Nov 27, 2026: Onboarding Writer | Skill | built Oct 2 | Kept | 36 min/week confirmed Metrics Reporter | Grok Bot | built Oct 9 | Kept | 20 confirmed Expense Manager | Grok Bot | built Oct 9 | Kept | 40 confirmed Invoice Chaser | Grok Bot | built Oct 16 | Retired at 30 days: you rewrote 9 of 10 drafts Proposal Follow-up | Skill | built Oct 23 | Kept | 25 confirmed Client Health Reporter | Grok Bot | Idea | Priority 18 Your counts for Oct 30 to Nov 27: 21 scheduled runs, 1 ROUTINE FAILED (Expense Manager, Nov 13, card CSV not downloaded). 16 approvals answered: 10 Invoice Chaser, 4 Metrics Reporter, 2 Bot Generator.
사용 후
METRICS, Oct 30 to Nov 27, 2026 Hours saved per week: 2.0 (Expense Manager 40, Onboarding Writer 36, Proposal Follow-up 25, Metrics Reporter 20 minutes, all confirmed) Kept after 30 days: 4 of 5 (80%) Routine failure rate: 1 of 21 runs (4.8%): Expense Manager, Nov 13, card CSV not downloaded; reported, not run on old data Approvals per week: 4 (Invoice Chaser had 10 of 16 before it was retired) Next: Invoice Chaser failed because its drafts needed rewriting 9 times out of 10: the job varies too much for a routine. Build the top open row instead, Client Health Reporter (Priority 18), as one weekly routine with a freshness rule on all 3 dashboards.
구성
- SkillBot Generator SkillInterview a job, pick Skill, Bot or dot, and generate it with a trust audit.
- BotBot Generator BotA Grok Bot that designs focused Bots, skills and routines, and asks before it builds.
- DotBot Generator DotYour dot spots repeat work in Gmail and Slack and drafts the next automation.
- bot-generator-kit-playbook.md
Kit 소개
Turn a job you keep doing by hand into a focused, tested AI teammate in under an hour, then add one a week. The Bot Generator Kit interviews you about the job, decides whether it should be a Skill, a Grok Bot or a ChatGPT dot responsibility, generates the complete artifact, and audits it for approvals, stale data, broad triggers and secrets before anything runs. It is for anyone who wants AI teammates that get used and never send, pay or delete without asking.
What you get
- Bot Generator Skill for Claude, ChatGPT, Codex and Cursor: the six-question interview, the decision rules for which form to build, complete templates for a SKILL.md folder, a Grok Bot blueprint and a dot responsibility pack, an approval boundary library for eight action types, a routine design checklist, a 16-point trust audit rubric, and the worked Expense Manager example.
- Bot Generator Bot for the Grok Bot app: the Bot that builds Bots. Five skills (Job interview, Bot blueprint, Skill from a finished task, Routine trust audit, Template prep) and three routines (Friday backlog review, Monday 30-day check, monthly roster review), with a Notion backlog and roster. It creates a new Bot only after you approve.
- Bot Generator Dot for ChatGPT: makes your dot an automation scout that reads Gmail and a few Slack channels, keeps a ranked backlog with estimated time saved, and brings you one build-or-skip decision every Friday. Reading is free; messaging anyone needs your OK; installing and sharing are handed to you.
- The playbook, only in the bundle: the setup order, how the three pieces hand work to each other, the weekly rhythm, the four metrics that matter, and a 30-day plan from one automation to a small team of focused Bots.
All three use the same INTERVIEW, FORM DECISION, NEXT AUTOMATION and TRUST AUDIT blocks. A candidate your dot spots on Friday becomes a Bot design in Grok or a SKILL.md in Cursor without retyping anything.
Who it is for
- Founders, operators and consultants who have tried an AI agent that went unused or did something it should not have.
- People who already work in ChatGPT, Claude, Cursor, Codex or the Grok Bot app and want a repeatable way to hand off work.
- Team leads who will share Bots with colleagues and need them free of keys, internal links and customer data.
What it will not do
- It will not install, enable, run or share anything for you. Each piece hands you the artifact, the install steps and a test plan with safe inputs.
- It will not send, publish, purchase, delete or change production systems without your approval.
- It will not build a "general helper" or a routine that listens to every new message.
- It will not automate judgment work such as hiring decisions, reviews or legal sign-off.
- It will not invent platform features. Every step uses menus that exist today.
Bought separately
The three components cost $19 each ($57). The bundle is $39, saves you $18, and adds the playbook, which only the bundle includes.
리뷰
아직 리뷰가 없어요.
Bundle 더 보기
Content Factory Kit
One long piece in, a week of on-voice posts out, scheduled in Notion and reviewed weekly.
Design Studio Kit
Brand kits, scored critiques and launch checks so AI-built UI stops looking generic.
Engineering Outer Loop Kit
Run your team's outer loop: daily triage, CI verdicts, release notes and postmortems.