agentleFS
andrerodriguesopsandrerodriguesops/gtm-engine/.claude/skills/setup/SKILL.md

setup

Skill0 starsChanged 9 days agoMITView on GitHub

Guided first-run setup for this GTM engine. Use when the user says "set me up", "get started", "walk me through setup", "where do I start", "continue setup", or when CLAUDE.md's First run rule fires (the canon still holds placeholders). Walks the user through SETUP.md one step at a time as a conversation, does the writing for them, verifies each step's check, and records progress in SETUP-PROGRESS.md so a later session resumes where this one stopped.

191 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use add-voice-transcription, add-discord and 10 more.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • juancruzmunozalbeloSkill

    add-voice-transcription

    Add voice message transcription to NanoClaw using OpenAI's Whisper API. Automatically transcribes WhatsApp voice notes so the agent can read and respond to them.

    Also in 5 projects that use this

  • MicrosoftSkill

    add-discord

    Add Discord bot channel integration via Chat SDK.

    Also in 11 projects that use this

  • MicrosoftSkill

    add-slack

    Add Slack channel integration via Chat SDK.

    Also in 11 projects that use this

  • juancruzmunozalbeloSkill

    add-telegram-swarm

    Add Agent Swarm (Teams) support to Telegram. Each subagent gets its own bot identity in the group. Requires Telegram channel to be set up first (use /add-telegram). Triggers on "agent swarm", "agent teams telegram", "telegram swarm", "bot pool".

    Also in 7 projects that use this

  • +8 more from microsoft

The file

We split this skill into the rules it gives an agent

7 rules. 0 are common, 5 are distinctive, and 2 are very specific to this project.

  • 0

    Common

    Most of the 168 edited versions we read keep these.

  • 5

    Distinctive

    This author's own choices. Worth a look.

  • 2

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Read SETUP.md and SETUP-PROGRESS.md.

    line 18
  2. Distinctive:

    If nothing is ticked, say in two or three sentences what they are about to build, that the whole thing takes about an hour to a first …

    line 19
  3. Distinctive:

    If some steps are ticked, say which step is next and what it ends with, then start it.

    line 22
  4. Distinctive:

    One step at a time, one question at a time: Never paste the whole of SETUP.md back at them.

    line 26
  5. Very specific:

    Never invent a fact about their company: If they don't know an answer yet, write it as an open row in NEEDS.md and keep going.

    line 34
  6. Distinctive:

    Commit with a message that says which step was completed.

    line 51
  7. Distinctive:

    When step 3a is done (first call scored by hand), the engine is usable.

    line 65

“Common” is measured: kept by at least 6 in 10 of the 168 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Setup — walk the user through it

The steps, their order and their checks live in SETUP.md. This skill is the method for walking someone through them. Do not restate the steps here or anywhere else.

Start

  1. Read SETUP.md and SETUP-PROGRESS.md.
  2. If nothing is ticked, say in two or three sentences what they are about to build, that the whole thing takes about an hour to a first scored call, and that they can stop at any step and pick up later by saying "continue setup". Then start step 0.
  3. If some steps are ticked, say which step is next and what it ends with, then start it.

How to run each step

  • One step at a time, one question at a time. Never paste the whole of SETUP.md back at them. Ask, wait, write, move on.
  • You write, they talk. The user never opens a file. Turn their answers into the files the step names. Show a short summary of what you wrote, not the whole file.
  • Interview before writing. In step 1, cover all three parts (what they sell, who outbound targets, which tools) before writing anything. Ask for examples: a real deal they won, a real objection, the price they actually quote. Push back on vague answers ("mid-market companies") until they are specific enough to score a call against.
  • Never invent a fact about their company. If they don't know an answer yet, write it as an open row in NEEDS.md and keep going. Arithmetic from a rule they stated ("annual is two months free" → annual = 10 × monthly) is written, and flagged in NEEDS.md for them to confirm.
  • Ask who signs and who champions separately. "The buyer" is often two people: the one who feels the pain and the one who signs. company-context has a row for each.
  • Steps you cannot do for them (connecting a connector, creating a CRM property in the UI, adding a GitHub secret, creating a Routine): tell them exactly where to click, one action per line, then wait until they say it's done. If they can't do it yet (no paid plan, no access), record it in NEEDS.md and move on; mark the box [~] so the first-run prompt stops asking.
  • Steps you can do for them (writing files, running the checks, calling a connected CRM to read stage IDs): do them, then show the result.
  • Run the check. Most steps in SETUP.md end with a Check. Run it or have them run it. A step is done only when its check passes. If it fails, fix the cause before moving on.

After each step

  1. Tick the step in SETUP-PROGRESS.md with today's date, and note anything left open.
  2. Commit with a message that says which step was completed.
  3. Say in one line what now works that didn't before, then offer the next step. If they want to stop, tell them to say "continue setup" next time.

Tools they don't have

STACK.md is a comparison, not a mandate. If they use a different CRM, call recorder or scheduler from the defaults, adapt the step to their tool and record the choice where step 1 says to. If their tool has no Claude connector or API that can do the step, say so plainly, add it to NEEDS.md, mark the box [~], and skip to the next step that doesn't depend on it. With no CRM, step 3 still runs: score calls with sales-call-scorer alone (SETUP.md §3).

Done

When step 3a is done (first call scored by hand), the engine is usable. Steps 4 and 5 automate it. When every box is ticked, tell them setup is complete and what to say day to day: "prep me for [company]", "process this call", "how's the pipeline".

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.