agentleFS
GerardoRdz96GerardoRdz96/claude-skills/skills/onboard/SKILL.md

onboard

Skill0 starsChanged 4 months agoMITView on GitHub

Use ONLY inside a freshly cloned aios-starter-kit clone (the repo has aios-intake.md and a CLAUDE.md with {{...}} placeholders) — NEVER in Servy or any already-onboarded AIOS. Day 1 of an AIOS install, when someone says "set me up", "onboard me", "let's get started", "fill in my AIOS", or has just cloned the kit. Combined wizard — runs the 7-question intake AND scaffolds the Day-1 file set at the end. Idempotent — re-run any time after editing aios-intake.md.

  • Reads credentials

165 other projects use a version of this, 9 of them edited.

Adopted by

Most recently taken up by

Used with

Projects with it also use soak-test, team-live-ops 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.
  • lilangMaxSkill

    soak-test

    Generate a soak test protocol for extended play sessions. Defines what to observe, measure, and log during long play sessions to surface slow leaks, fatigue effects, and edge cases that only appear after sustained play. Primarily used in Polish and Release phases.

    Also in 9 projects that use this

  • lilangMaxSkill

    team-live-ops

    Orchestrate the live-ops team for post-launch content planning: coordinates live-ops-designer, economy-designer, analytics-engineer, community-manager, writer, and narrative-director to design and plan a season, event, or live content update.

    Also in 9 projects that use this

  • HK-hubSkill

    teach-impeccable

    One-time setup that gathers design context for your project and saves it to your AI config file. Run once to establish persistent design guidelines.

    Also in 19 projects that use this

  • CoralGameSkill

    project-stage-detect

    Automatically analyze project state, detect stage, identify gaps, and recommend next steps based on existing artifacts.

    Also in 15 projects that use this

  • CoralGameSkill

    team-level

    Orchestrate level design team: level-designer + narrative-director + world-builder + art-director + systems-designer + qa-tester for complete area/level creation.

    Also in 11 projects that use this

  • +5 more from lilangMax
  • +2 more from CoralGame

The file

We split this skill into the rules it gives an agent

12 rules. 0 are common, 6 are distinctive, and 6 are very specific to this project.

  • 0

    Common

    Most of the 153 edited versions we read keep these.

  • 6

    Distinctive

    This author's own choices. Worth a look.

  • 6

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    If the user has already onboarded and wants to refresh: still run, but skip questions already answered (idempotent).

    line 15
  2. Very specific:

    If the user wants to add a new connection: that's not onboarding — point them at connections.md to edit directly, or schedule a /level-up …

    line 16
  3. Very specific:

    Read aios-intake.md. Check which Q1-Q7 sections have content vs.

    line 22
  4. Very specific:

    Ask one at a time. Write each answer into aios-intake.md as you go (so the user can resume if interrupted).

    line 30
  5. Distinctive:

    Ask for two samples. One email, one post.

    line 40
  6. Distinctive:

    When the user runs the closing prompt ("what should I focus on this week?"), respond using only the new context files.

    line 82
  7. Distinctive:

    The 7-question cap is non-negotiable: Don't add Q8 in conversation.

    line 91
  8. Distinctive:

    Voice paste cannot be skipped: If the user types samples mid-chat, refuse and tell them to paste from real writing.

    line 92
  9. Very specific:

    One-shot scaffold: After Step 2 ends, write Step 3 files in a single batch.

    line 93
  10. Distinctive:

    Closing screen is three lines: Not a menu.

    line 95
  11. Very specific:

    No extra skills generated: Don't scaffold /today, /draft, /connect, etc.

    line 96
  12. Very specific:

    No .env writes: Don't ask for API keys on Day 1.

    line 98

“Common” is measured: kept by at least 6 in 10 of the 153 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

What this skill does

Single combined wizard. Reads or writes aios-intake.md (the canonical intake), conducts the 7-question interview if the file isn't filled, then scaffolds the Day-1 file set inline at the end of the run. No separate /scaffold-from-intake skill — this is one flow.

The wow moment: at the end, suggest the closing prompt "Try this — ask me: what should I focus on this week?" The user runs it once. That's the wow. There's no /today skill to save — the prompt itself plants the Mindset framework (Default Shift) for them to internalize.

When NOT to run this

  • If the user has already onboarded and wants to refresh: still run, but skip questions already answered (idempotent).
  • If the user wants to add a new connection: that's not onboarding — point them at connections.md to edit directly, or schedule a /level-up Phase 2 walk.

Execution

Step 1: Read the intake

Read aios-intake.md. Check which Q1-Q7 sections have content vs. [Your answer here] placeholders.

  • All filled → skip Step 2, jump to Step 3 (scaffold).
  • Some filled → ask the user: "I see Q1, Q3, Q4 are answered. Want to fill the rest now, or scaffold from what's there?" Their call.
  • None filled (fresh clone) → run Step 2 conversationally.

Step 2: The interview (7 questions, hard cap)

Ask one at a time. Write each answer into aios-intake.md as you go (so the user can resume if interrupted).

Q1 — Who are you, what do you sell, who do you sell it to? Identity, offer, ICP. One paragraph each is fine.

Q2 — Paste 1-2 things you've written recently. Don't edit them. This is the only question with a hard rule. Voice samples MUST be pasted, not typed mid-conversation. If the user starts typing fresh prose, refuse:

"Stop — paste it raw. If you type it here while we're talking, the sample is already shaped by our conversation. Open your last email or LinkedIn post in another tab and paste the unedited text. This is the one rule I can't bend."

Ask for two samples. One email, one post. Or two of either.

Q3 — What are your 2-3 biggest priorities for the next 90 days? Quarterly priorities. Push back if they say "grow my business" — make them name a number, a deadline, or a deliverable.

Q4 — Where does revenue actually land, and where is it tracked? Multiple answers OK. Map to Tier-1 Domain 1 (Revenue/Financials).

Q5 — Where do you talk to customers, your team, and the outside world day-to-day? Email (Gmail/Outlook), Slack/Teams/Discord, DMs. Map to Domains 2 + 4.

Q6 — Where do meeting recordings, notes, and important docs live? Map to Domains 6 + 7.

Q7 — What's the one task that eats your week, and where do you currently track work? Capture top_pain (used by /level-up Day-14) + Domain 5 (tasks).

Domain 3 (Calendar) is auto-inferred from Q5: Gmail → Google Cal; Outlook → Outlook Cal. Confirm in Step 3.

Step 3: Scaffold the Day-1 file set

Once the intake is complete, generate these files (or update if re-running). Back up originals to archives/intake-{YYYY-MM-DD-HHMM}/ if any exist.

  1. context/about-me.md — from Q1 (identity, role) + Q7 (top_pain). One short paragraph each.
  2. context/about-business.md — from Q1 (offer, ICP) + Q4 (revenue model). One paragraph.
  3. context/priorities.md — from Q3. Numbered list, one line per priority.
  4. references/voice.md — from Q2. Paste samples verbatim with a short header explaining their use ("Match this register when drafting; don't fake voice on external content without showing me first").
  5. connections.md — populate the 7-row table from Q4-Q7 answers. Each row gets mechanism: not yet connected, auth: —, last checked: —. The user wires connections on Day 2.
  6. CLAUDE.md — fill all {{...}} placeholders. Substitute the user's name, stated priority, voice register summary, and a brief connections summary.

Step 4: The closing screen

Print one screen. Three lines max:

✓ Day 1 done. Your AIOS knows who you are, what you sell, what matters this quarter, and how you sound.

Today: ask me — "what should I focus on this week?"
Tomorrow: pick one tool from connections.md and wire it up (manual MCP install or write a small API script + save references/{tool}-api.md).
Day 7: run /audit to see your score.

When the user runs the closing prompt ("what should I focus on this week?"), respond using only the new context files. Hit:

  • 3-bullet priority list, in their voice register from Q2
  • Each bullet ties back to a stated 90-day priority from Q3
  • Final line: "If I had to pick one thing for Monday, it'd be [X], because [reason from priorities]. Want me to draft the first email? And — where could the Default Shift apply here? To what extent could AI be leveraged on this task?"

The Default Shift question seeds the Mindset framework before /level-up formally introduces it on Day 14.

Critical implementation rules

  1. The 7-question cap is non-negotiable. Don't add Q8 in conversation.
  2. Voice paste cannot be skipped. If the user types samples mid-chat, refuse and tell them to paste from real writing.
  3. One-shot scaffold. After Step 2 ends, write Step 3 files in a single batch. No multi-turn confirmation. The user iterates by editing aios-intake.md and re-running.
  4. Idempotent. Re-running with an edited intake refreshes context files; backs up originals to archives/intake-{ts}/. Skips questions already answered unless the user wants to revise.
  5. Closing screen is three lines. Not a menu.
  6. No extra skills generated. Don't scaffold /today, /draft, /connect, etc. The kit ships 3 skills; the user authors more via /level-up.
  7. Read-only on references/3ms-framework.md. It already ships in the kit. Don't overwrite.
  8. No .env writes. Don't ask for API keys on Day 1. Connections come Day 2.

Verification (for the implementer)

  • Cold-test: clone a fresh kit, run /onboard, fill 7 answers, scaffold runs, ask the wow prompt, response cites Q1 + Q3 + Q7 specifically. Generic = fail.
  • Idempotency: re-run /onboard with one Q3 priority changed. Expected: only context/priorities.md and CLAUDE.md's priority section update; backup created in archives/intake-{ts}/.
  • Voice rejection: type a sample mid-chat. Expected: skill refuses, asks for paste.

Adapted from The Three Ms of AI™ © 2026 Nate Herk. The Mindset language used in the closing screen comes from references/3ms-framework.md.

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.