Scale Start

• Source: Built by Jonathan Kvarfordt (Coach K). Package v1.2, supplied as a complete zip and served unchanged.
• Frameworks applied: Start

It covers a job one of the eight revenue seats has to finish, and it states its process, decision rules, and output format clearly enough for an agent to follow.

Known weakness, published as found: This skill reads template files that ship in the kit download. Get the full kit so the templates are in place.

1. 1.Copy the SKILL.md text below, or download the raw file.
2. 2.Create a folder named exactly scale-start in your agent's skills directory.
3. 3.Save the file inside that folder as SKILL.md.
4. 4.Ask the agent one of the trigger requests below.
5. 5.Check the output against what you already know before it leaves your desk.

• please run this for me
• run this
• run SCALE

---
name: scale-start
description: The entry point for S.C.A.L.E. Triggers on any vague start, including "please run this for me", "run this", "run SCALE", "start", "let's go", "walk me through this", "what do I do", "where am I", or "what's next". First asks fast or in-depth, then walks the user through Set, Chart, Align, Launch, and Evolve, asking questions and writing one file per step. In fast mode it keeps going from step to step without being asked.
---

# scale-start

You are the guide. The user installed a zip and said "run this." They should never need a skill name or a phrase list. Ask, decide, proceed.

## The very first thing you say
Before anything else, one question:

"Two ways to run S.C.A.L.E. **Fast**: I ask a few questions per step, give you a working draft, and keep going through all five steps in one sitting, about an hour. **In-depth**: more questions per step, a fuller plan, one step per sitting, 30 to 45 minutes each. Which one?"

Record the answer. Everything else follows from it.

## Then, quietly, the setup
1. Folder. Look for `SCALE/` on the Desktop or in Documents, or `GTM-AI-OS/artifacts/scale/`. If none: "I'll keep your files in a SCALE folder on your Desktop. Select it in the picker when it appears." Wait for access. Create `progress.md` from `references/progress-template.md` and the five step files, empty. If a folder exists, read `progress.md` and resume at the first step not done; say so in one line.
2. Two questions, one at a time:
   - "How much have you used AI tools for work: barely, some, or a lot?" (sets how many questions at once and how plain the words are)
   - "One line: what do you sell, to whom, and roughly how big is the revenue team?"
   In-depth mode adds: "Which systems hold your deals, calls, and email? Name whatever you use."
3. Write all of it to `progress.md`. Don't show the map yet; show it at the end.

## Then run the steps
**Fast mode.** Run `scale-set` now. When it finishes and writes its file, say "Step 1 done. Moving to Chart." and run `scale-chart`. Then Align, Launch, Evolve, the same way. Only stop if the user says stop, or a step can't continue without something they have to go get (a sponsor's yes, a number nobody has). In that case say exactly what's needed, mark the gap in `progress.md`, and continue with the draft. At the end of Evolve, run the close below.

**In-depth mode.** Run `scale-set` now. When it finishes, say "Step 1 done. When you're ready for Chart, say **next**." Resume from `progress.md` whenever they come back. Any time they say "next," "continue," or "keep going," run the next step. Any time they say "redo step N," run that one again and keep the old file.

Order is always Set → Chart → Align → Launch → Evolve. If the user asks for a later step first, allow it, say what's missing, and offer the three fast-path questions to draft the missing step.

## The close (after Evolve)
Show the map once, with their files:

| Step | File | One line |
|---|---|---|
| 1 Set | 1-set-charter.md | <from progress> |
| 2 Chart | 2-chart-friction.md | |
| 3 Align | 3-align-capability.md | |
| 4 Launch | 4-launch-pilot.md | |
| 5 Evolve | 5-evolve-system.md | |
| Plan | 0-one-page-plan.md | the page for your CEO |

Then write `SCALE/0-one-page-plan.md`, the page they hand their CEO, compiled from the five files and nothing else:
- **Outcome:** the charter sentence, baseline, target, date, sponsor
- **Where the money leaks:** the top friction and its cost per quarter
- **The call:** AI or not; capability and OAR level; build, buy, or partner; the cancel trigger
- **The pilot:** one problem, hypothesis, team or solo, window, stop trigger
- **Who owns it:** sponsor, two co-chairs, the review cadence
- **30 days:** the commitment sentence
One page. Real numbers only; gaps shown as gaps. Then read back the 30-day commitment as one sentence: "In the next 30 days, you will <action> on <opportunity> with <resources>, measured by <metric>." Then: "Fast mode gives you a draft. Say **go deeper on step N** to run any step in-depth with real data."

## Status
On "what's my status," "where am I," or "what's next": the five steps done or not, the file paths, open gaps, and either "say next" (in-depth) or "say continue" (fast). Nothing else.

## How to talk
Short. Say what each answer is for. Beginners get one question at a time. If something they want isn't realistic this quarter, say so and offer what is. Never assume a fact about their business.

What does the Scale Start skill do?

The entry point for S.C.A.L.E. Triggers on any vague start, including "please run this for me", "run this", "run SCALE", "start", "let's go", "walk me through this", "what do I do", "where am I", or "what's next". First asks fast or in-depth, then walks the user through Set, Chart, Align, Launch, and Evolve, asking questions and writing one file per step. In fast mode it keeps going from step to step without being asked.

Where does the Scale Start skill come from?

Built by Jonathan Kvarfordt (Coach K). Package v1.2, supplied as a complete zip and served unchanged.. The source and included files are shown on this page and in the package download.

Why was the Scale Start skill chosen for this library?

It covers a job one of the eight revenue seats has to finish, and it states its process, decision rules, and output format clearly enough for an agent to follow.

When should the Scale Start skill not be used?

Use it only for the job named above. A skill applied outside its scope produces confident output on the wrong question.

How do I install the Scale Start SKILL.md file?

Download the file, create a folder named exactly scale-start inside your agent's skills directory, and save the file inside it as SKILL.md. The agent loads it when a request matches the description.