agentleFS
DINAKAR-SDINAKAR-S/company-ai-os/.claude/skills/onboard/SKILL.md

onboard

Skill2 starsChanged 3 months agoMITView on GitHub

Onboard ANY company (or founder/freelancer/solo owner) into this AI OS. Interviews them about their business, tools, and APIs; researches available MCP servers and integrations; then GENERATES their custom entity notes, routing table, department agents, skills, and style guide. Use on first run after cloning, or when the user says "onboard", "set this up for my company", or "/onboard".

171 other projects use a version of this.

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 10 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 10 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 18 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 12 projects that use this

  • CoralGameSkill

    playtest-report

    Generates a structured playtest report template or analyzes existing playtest notes into a structured format. Use this to standardize playtest feedback collection and analysis.

    Also in 12 projects that use this

  • HK-hubSkill

    arrange

    Improve layout, spacing, and visual rhythm. Fixes monotonous grids, inconsistent spacing, and weak visual hierarchy. Use when the user mentions layout feeling off, spacing issues, visual hierarchy, crowded UI, alignment problems, or wanting better composition.

    Also in 13 projects that use this

  • +5 more from lilangMax
  • +1 more from CoralGame

The file

We split this skill into the rules it gives an agent

4 rules. 0 are common, 3 are distinctive, and 1 is very specific to this project.

  • 0

    Common

    Most of the 159 edited versions we read keep these.

  • 3

    Distinctive

    This author's own choices. Worth a look.

  • 1

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Ask in small batches (3–5 questions at a time), adapt follow-ups to their answers:

    line 14
  2. Very specific:

    Run /lint to verify the generated OS is consistent.

    line 65
  3. Distinctive:

    Every integration claim is web-verified with a date.

    line 76
  4. Distinctive:

    Delete sample content you replace.

    line 78

“Common” is measured: kept by at least 6 in 10 of the 159 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

Onboard — turn this template into THEIR company's OS

You are the onboarding guide. The goal: a stranger clones this repo, runs /onboard, answers questions, and walks away with a working OS custom-built for their business. Work in four phases. Announce each phase as you enter it.

Phase 1 — Interview (ask, don't assume)

Ask in small batches (3–5 questions at a time), adapt follow-ups to their answers:

  1. Who are you? Company/personal brand name, what you do, team size (solo founder and freelancer are valid answers — the OS scales down fine).
  2. Functions. Which of these exist for you: HR/people, engineering/product, operations, marketing, finance, sales/clients, support — plus anything unusual (inventory, dispatch fleet, clinics, classrooms…). One entity note each.
  3. Tools per function. For each function, what do they run on? Examples to prompt them: Meta/Google Ads, social schedulers, CRM, payroll system, inventory/POS, accounting, e-commerce, helpdesk, WhatsApp Business, email.
  4. Where files live. Real folders (paths), drives, or "everything is in SaaS".
  5. Writing channels. Which do they write in — email/Gmail, WhatsApp, LinkedIn, Instagram? Ask for 1–2 real (non-sensitive) sample messages per channel to learn their voice.
  6. Boundaries. What's sensitive (salaries, customer PII, credentials)? Confirm the rule: referenced, never copied into the vault.

Phase 2 — Research integrations (never guess)

For EACH tool they named:

  1. Web-search for an official MCP server first, then an official API.
  2. Determine the connection path, best-first:
    • MCP server exists → exact install command (claude mcp add …) and what auth it needs
    • API only → where to get the API key (exact console URL), which scopes
    • Neither → manual export/CSV workflow into raw/
  3. Produce an Integration Plan table: tool | function | connect via | steps to get access | date checked. Everything web-verified, nothing from memory — note the date.
  4. Walk them through connecting ONE integration end-to-end as the proof; list the rest as checklist items. API keys go in env vars / agent settings — never into vault files. (Full patterns: docs/INTEGRATIONS.md.)

Phase 3 — Generate their OS

With their answers, create (vault writes need no confirmation):

  1. entities/Company.md — rewritten for them (delete Acme sample data).
  2. One entity note per function they actually have (delete unused sample departments, or repurpose them). Each carries real path:, tools, and the integration plan rows.
  3. entities/Routing.md — rebuilt with their real names/aliases/paths.
  4. entities/Style Guide.md — their voice, learned from the samples they gave: rules + one worked example per channel (email, WhatsApp, LinkedIn…).
  5. Custom agents in .claude/agents/ — one per major function, cloned from the shipped examples, rescoped to their entity note, tools, and boundaries. (Inventory company → inventory-manager agent wired to their POS MCP; marketing agency → ads-analyst wired to Meta/Google APIs.)
  6. Custom skills in .claude/skills/ — one per recurring ritual they described (weekly report, campaign launch checklist, stock reorder check…).
  7. Update index.md and Home.md — their map, their checklist.

Phase 4 — Handoff

  1. Run /lint to verify the generated OS is consistent.
  2. Log: ## [date] ingest | onboarded <Company> → N entities, M agents, K skills, integration plan.
  3. Show them the daily loop: /session-start → work → /eod → /session-handoff, and the learning loop (learnings.md / mistakes.md — see CLAUDE.md §7).
  4. End with their personal "first week with the OS" plan: 3–5 concrete tasks (e.g. "ingest your price list", "connect the second integration", "first payroll checklist run").

Rules

  • Interview before generating — never scaffold from assumptions.
  • Every integration claim is web-verified with a date. Never guess an API or price.
  • Scale to size: a freelancer might need 3 entities and 1 agent, not 6 and 4.
  • Delete sample content you replace; the OS should contain only THEIR company when done.

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.