onboard

Project Onboarding

This skill brings the full team up to speed on an existing project they've never seen. Each persona assesses the project from their domain, producing a baseline understanding with gaps, risks, and recommendations. The lens for every assessment is user value: is this project delivering value to its users, and what's getting in the way?

Model Selection

When spawning agents, pass model: explicitly per _shared/orchestration.md (Agent Model Selection). For this skill:

• IT Architect (leads component identification + tier proposal): opus — needs reasoning to draft the component list and propose tiers from observable signals
• Other personas: sonnet

Onboarding runs before COMPONENTS.md exists, so no tier modulation applies. Use the assignments above as written.

Process

Step 0: Gather Project Context

Before spawning agents, collect what's available:

Project structure:

find . -type f -name "*.md" -o -name "*.json" -o -name "*.yaml" -o -name "*.yml" -o -name "*.toml" -o -name "*.tf" -o -name "*.py" -o -name "*.ts" -o -name "*.js" | head -100

Documentation (read, not just list):

cat README.md 2>/dev/null
cat CLAUDE.md 2>/dev/null
cat CONTRIBUTING.md 2>/dev/null
cat DEVELOPMENT.md 2>/dev/null
cat SECURITY.md 2>/dev/null
cat CHANGELOG.md 2>/dev/null | head -50
ls docs/ adrs/ docs/adrs/ runbooks/ 2>/dev/null

Git history — activity, contributors, releases:

git log --oneline -20 2>/dev/null
git shortlog -sn --since="3 months ago" 2>/dev/null
git tag -l --sort=-version:refname | head -10 2>/dev/null

Stack — packages, toolchain versions:

cat package.json 2>/dev/null | head -40
cat pyproject.toml 2>/dev/null | head -40
cat requirements.txt 2>/dev/null | head -20
cat go.mod 2>/dev/null | head -10
cat .tool-versions .nvmrc .python-version mise.toml 2>/dev/null

Infrastructure & service topology (read, not just list):

cat docker-compose*.yml 2>/dev/null
cat Dockerfile* 2>/dev/null
ls -la terraform/ infrastructure/ deploy/ ansible/ 2>/dev/null
cat .env.example 2>/dev/null