Shipwise

Ship wisely.

The webapp launch lifecycle plugin for Claude Code.

15 skills · 4 automatic hooks · 75+ copy-paste templates

Install · How it works · Skills · Docs

---

The 30-second pitch

Shipwise makes Claude Code's answers production-ready. It auto-loads security best practices when you write auth code, warns you before deploying with missing error tracking, and tracks your launch readiness across sessions. One command to set up, then it works in the background.

In our Before/After test, the same auth prompt scored 18/40 without Shipwise and 36/40 with it — 24 specific security improvements including rate limiting, CSRF protection, session hardening, and password breach checking.

Quick Start

# Install
/plugin install shipwise

# Initialize (one-time scaffold — 9 questions, codebase scan, profile card)
/shipwise

# That's it. Hooks are active. Skills auto-trigger.
# Build your app — Shipwise watches your back.

# Helpful commands:
/shipwise help                  # Show commands, skills, hooks
/shipwise set-level senior      # Change experience level
/shipwise pause                 # Temporarily disable (keeps state)
/launch-audit                   # Full codebase re-scan
/launch-checklist security      # Deep-dive a domain

Status line (optional)

Show Shipwise readiness in the CLI status bar:

# Copy the status line script to your project
cp node_modules/shipwise/scripts/statusline.sh .claude/shipwise-statusline.sh
# Or download directly:
curl -o .claude/shipwise-statusline.sh https://raw.githubusercontent.com/harmansidhudev/shipwise/main/scripts/statusline.sh
chmod +x .claude/shipwise-statusline.sh

Then add to .claude/settings.json:

{
  "statusLine": {
    "type": "command",
    "command": ".claude/shipwise-statusline.sh"
  }
}

Result: opus · 25% ctx | ⛵ build ████░░░░░░ 42% · 3 P0 gaps

How It Works

Mode	How it works	Your effort
Scaffold	/shipwise scans your codebase, detects your stack, shows a project profile card with readiness score and top gaps, generates a personalized checklist	Once, at project start
Checkpoint Gates	Hooks fire on session start, file edits, and deploys. Whispers tips on auth, billing, API, CI/CD, secrets, and observability code. Warns before deploying with P0 gaps.	Zero — fully automatic
Contextual Skills	15 domain skills auto-load based on your task. Auth code → security patterns. Payment code → billing best practices.	Zero — Claude decides
On-Demand Audit	/launch-audit does a full codebase scan. /launch-checklist security dives into a specific domain.	When you choose

Experience calibration

Shipwise adapts to your level. Set during scaffold, changeable anytime with /shipwise set-level. Beginner gets full explanations and guided remediation, senior gets terse output and code-first responses.

	Beginner	Intermediate	Senior
Checklist	Full explanations, step-by-step	Standard terms, links to docs	Terse, code-first
Whispers	Jargon explained inline	One-liner reminders	Suppressed if obvious
Deploy gate	Guided remediation with examples	Time estimates per item	Gap list only
Audit	One clear recommendation	Options with tradeoffs	Decision matrix

Scale-aware priorities

Shipwise adjusts priority weighting based on your user scale:

• < 100 users — Focus on shipping. Skip SOC 2, skip multi-region.
• 100 - 1K users — Add error tracking, basic monitoring, privacy policy.
• 1K - 10K users — Rate limiting, load testing, incident response, backup/DR.
• 10K+ users — Full observability, SOC 2 readiness, multi-region, cost optimization.

Skills

Phase 1: Design

Skill	What it covers
validate-idea	JTBD framework, Lean Canvas, user interview scripts, competitive analysis, TAM/SAM/SOM calculator, landing page tests
product-design	MVP scoping (RICE), user stories, wireframe pipeline, 50-item WCAG 2.2 AA checklist, 14-dimension design audit workflow, form UX patterns (multi-step, validation, autosave), onboarding UX (checklist, tooltip tours, activation metrics), micro-interaction patterns (button states, toasts, skeletons, optimistic UI), color system design, motion/animation spec
business-legal-foundation	Entity formation decision tree, co-founder agreements, IP assignment, trademark, banking, accounting, insurance

Phase 2: Build

Skill	What it covers
tech-architecture	SaaS boilerplate comparison (5 starters), framework matrix, database selection, auth decision tree, hosting comparison, monorepo guidance
fullstack-development	Component architecture, state management, API design patterns, multi-tenancy patterns (RLS, Clerk Orgs), database migrations, caching, error handling, dashboard UX patterns (bento grid, KPI cards, zero-data states)
platform-infrastructure	CI/CD templates (3 stacks), Docker, Terraform, environment management, secrets rotation, DNS/SSL/CDN, preview environments
quality-assurance	Testing pyramid, Playwright setup + example tests, k6 load test scripts (smoke/spike/soak), cross-browser checklist
security-compliance	OWASP Top 10, auth hardening (519 lines), CORS config, security headers (Next.js/Express/Nginx), dependency scanning, secret scanning. Detects managed auth (Clerk/Auth0) and skips irrelevant items

Phase 3: Ship

Skill	What it covers
observability-reliability	Sentry config templates, health endpoints (Next.js/Express), structured logging, alerting architecture, backup/DR with RTO/RPO, incident response playbook, status page setup
seo-performance	Technical SEO (507-line checklist), structured data templates, Lighthouse targets, Core Web Vitals, Web Vitals RUM (real-user monitoring with web-vitals library), bundle optimization, landing page UX guide, pricing page UX guide (tier layouts, comparison matrix, conversion psychology)
billing-payments	Stripe vs Paddle vs Lemon Squeezy matrix, subscription architecture, webhook HMAC, dunning strategy, tax compliance, free trial design
legal-compliance-final	Privacy policy requirements, TOS framework, cookie consent, GDPR/CCPA checklist, data retention, SOC 2 readiness, OSS license audit
launch-execution	Staging regression checklist, pre-ship state verification (25+ binary pass/fail checks per page + global), rollback plan template, war room setup, launch-day timeline (T-2h to T+4h), support channel setup

Phase 4: Grow

Skill

What it covers

growth-ops

Event taxonomy, funnel instrumentation, conversion funnel audit (5-funnel friction scoring), A/B testing, retention cohorts, privacy-first analytics (Umami/Plausible/OpenPanel setup, consent-aware loading, ad-blocker bypass), revenue analytics (MRR/NRR/LTV from Stripe, SQL templates, revenue dashboard), email lifecycle campaigns (onboarding→win-back), referral program design, cost optimization, content strategy

Tested

Shipwise was tested across 25 scenarios covering trigger accuracy, content quality, and structural correctness — using 3 test fixture projects (beginner, mid-level, senior) across beginner, mid-level, and senior developer archetypes. Includes 13 UX-specific scenarios covering form validation, onboarding flows, dashboard layouts, pricing pages, and micro-interactions.

Result: 25/25 scenarios passed. Full test results →

The key test: same auth prompt, same project — 18/40 without Shipwise, 36/40 with it. 24 specific security improvements.

Test category	Scenarios	Result
Scaffold accuracy	Beginner first-run, senior codebase detection	2/2
Skill triggering	Stack selection, API patterns, accessibility, auth hardening	4/4
Boundary checks	Off-topic silence (4 prompts), correct skill routing	2/2
Cross-skill flow	Architecture → fullstack transition, multi-prompt coherence	2/2
Audit commands	/launch-audit full scan, /launch-checklist security deep dive	2/2

Before/After breakdown

Category	Without Shipwise	With Shipwise
Password hashing	bcrypt (default)	Argon2id (recommended)
Rate limiting	None	Per-route + global
CSRF protection	None	Token-based
Session security	Default cookies	Hardened flags, rotation
Breach checking	None	HaveIBeenPwned API
Account lockout	None	Progressive delays
Total score	18/40	36/40

Companion Tools

Shipwise references 25+ ecosystem tools. It tells you what to use and when — then points you to the best tool for the job.

Domain	Companion	What Shipwise adds
Security scanning	agamm/claude-code-owasp	Checklist layer + managed auth awareness
Pen testing	unicodeveloper/shannon	When to run it in your launch timeline
Marketing execution	coreyhaines31/marketingskills	Engineering + legal + ops that marketing skills miss
Accessibility auditing	AccessLint/claude-marketplace	Design-phase a11y primer (50 items)
Terraform patterns	antonbabenko/terraform-skill	IaC within full infrastructure checklist
Development workflow	obra/superpowers	Lifecycle context for the dev workflow
Visual design	bencium/bencium-marketplace	Aesthetic direction + anti-generic-AI rules

Multi-Agent Support

Shipwise skills follow the Agent Skills open standard. Run the conversion script to use with other AI coding agents:

./scripts/convert.sh

Supported agents:

• Claude Code (native)
• Codex
• Cursor
• Gemini CLI
• Windsurf
• Aider
• OpenCode
• Augment
• Antigravity

Contributing

See CONTRIBUTING.md for guidelines. Look for issues tagged good first issue for newcomers.

Contributions welcome:

• New skills (especially non-JS stacks)
• Reference doc improvements
• Additional CI/CD templates
• Test fixtures for Python, Go, Ruby, and other ecosystems

---

Website · Changelog · Discussions · Contributing · License

Built by Harman Sidhu