← ClaudeAtlas

operator-onboardinglisted

First-run setup for AI-PM Operator. Detects a fresh install, infers the user's business from the surrounding repo, asks a short set of questions, then writes business.json / .env / team.json and tailors every skill to the user's business. Use on first launch, when config is missing, or when the user says 'set up', 'onboard', 'configure', 'get started', or '/onboard'.
mohitkhandelwal242/ai-pm-operator · ★ 0 · AI & Automation · score 65
Install: claude install-skill mohitkhandelwal242/ai-pm-operator
# AI-PM Operator — Onboarding Your job: take someone who just pulled this repo and get them from zero to a working, **business-tailored** AI-PM Operator in one short conversation. They could be a PM at any kind of company (B2B SaaS, consumer app, marketplace, e-commerce, fintech, dev tools…). Nothing here is specific to any one business — you learn theirs and configure accordingly. **Run this when:** `business.json` is missing, `.env` is missing, the user asks to "set up / onboard / configure", or `--reconfigure` is passed. --- ## Step 0 — Detect state ```bash ls -la .env business.json team.json 2>/dev/null ``` - If `.env` and `business.json` both exist and `--reconfigure` was NOT passed → tell the user they're already set up, show `/operator-onboarding --reconfigure` to redo, and stop. - Otherwise continue. Copy templates if the working files don't exist yet: ```bash [ -f business.json ] || cp business.json.template business.json 2>/dev/null [ -f .env ] || cp .env.example .env 2>/dev/null ``` --- ## Step 1 — Infer the business from the repo (do this BEFORE asking) Scan the surrounding working directory so your questions are smart, not generic. Don't dump raw output — infer. - Tech stack & platforms: look for `package.json`, `requirements.txt`, `go.mod`, `build.gradle`/`AndroidManifest.xml`, `Podfile`/`*.xcodeproj`, `composer.json`, `Dockerfile`. - Product signals: read `README.md`, `package.json` (name/description), landing copy, `/docs`, app store config — infer pr