claude-code-standards-kit / templates
CLAUDE.md2 starsChanged 8 months agoMITView on GitHub
Universal template — Claude Code Standards Kit by SQWR Studio. Instructions: replace all [TO FILL IN] placeholders and remove irrelevant sections. Plugin mode: If using the SQWR plugin (.claude/hooks.json or via .claude-plugin/), hooks and slash commands are auto-discovered. Slash commands available: /new-feature, /pre-deployment, /monthly-review, /full-audit, /verify-project. [YOUR NAME] — [your role, e.g.: founder / developer / student] This project belongs to: [Your studio / your company / personal] Full context: - Identity: [KIT_PATH]/IDENTITY-TEMPLATE.md (fill in with your information) Name: [TO…
In this repository11 other files it gives its agents.
- global/CLAUDE.mdCLAUDE.md
- audit-runnerSkill
- auto-fixSkill
- brainstormSkill
- compliance-checkSkill
- contract-lookupSkill
- monthly-reviewSkill
- new-featureSkill
- pre-deploymentSkill
- project-setupSkill
- risk-scoreSkill
The file
Files
[PROJECT NAME] — AI Contract
Universal template — Claude Code Standards Kit by SQWR Studio. Instructions: replace all
[TO FILL IN]placeholders and remove irrelevant sections.Plugin mode: If using the SQWR plugin (
.claude/hooks.jsonor via.claude-plugin/), hooks and slash commands are auto-discovered. Slash commands available:/new-feature,/pre-deployment,/monthly-review,/full-audit,/verify-project.
Who you are
[YOUR NAME] — [your role, e.g.: founder / developer / student]
| Contact | Value |
| [your@email.com] | |
| Work email | [pro@yourstudio.com] |
This project belongs to: [Your studio / your company / personal]
Full context:
- Identity:
[KIT_PATH]/IDENTITY-TEMPLATE.md(fill in with your information)
This project
Name: [TO FILL IN] Description: [TO FILL IN — 1-2 sentences max] Production URL: [TO FILL IN or N/A] Status: [In development / In production / Archived]
Stack:
- [TO FILL IN — e.g.: Next.js 16 + TypeScript + Tailwind CSS + Supabase]
Deployment:
- [TO FILL IN — e.g.: Vercel (auto from main)]
Architecture
[TO FILL IN — key folder structure]
src/
├── app/ → Next.js App Router routes
├── components/ → Reusable components
├── lib/ → Utilities, config, helpers
└── types/ → Shared TypeScript typesCritical files:
| File | Role |
| [TO FILL IN] | [role] |
| [TO FILL IN] | [role] |
Active contracts
Include the contracts relevant to this project. Source files:
docs/contracts/Check [x] the ones active for this project. Read each checked contract before working on the relevant domain.
Core (check all that apply to your stack)
CONTRACT-TYPESCRIPT.md— Strict typing, noany, SOLIDCONTRACT-SECURITY.md— OWASP Top 10, NIST, secrets managementCONTRACT-TESTING.md— Test pyramid, TDD, coverage ≥80%CONTRACT-CICD.md— GitHub Actions, DORA metrics, Conventional Commits
Web stack
CONTRACT-NEXTJS.md— App Router, SSR, Server Components, CWVCONTRACT-SUPABASE.md— RLS, auth, migrations, NIST SP 800-63BCONTRACT-VERCEL.md— Deployment, env vars, rollbackCONTRACT-PERFORMANCE.md— LCP ≤2.5s, INP ≤200ms, CLS ≤0.1CONTRACT-ACCESSIBILITY.md— WCAG 2.1 AA, EAA (EU law since June 2025)CONTRACT-SEO.md— SSR, metadata, JSON-LD, Core Web VitalsCONTRACT-DESIGN.md— Gestalt, typography, colour tokens, TailwindCONTRACT-OBSERVABILITY.md— Structured logging, Sentry, RUMCONTRACT-RESILIENCE.md— Circuit breaker, retry, graceful degradation
AI features
CONTRACT-AI-PROMPTING.md— System prompts, few-shot, model selectionCONTRACT-ANTI-HALLUCINATION.md— Real data only, RAG, context poisoning
Mobile
CONTRACT-IOS.md— SwiftUI, HIG, Dark Mode, touch targets ≥44ptCONTRACT-ANDROID.md— Jetpack Compose, Material 3, TalkBack
Specialized
CONTRACT-EMAIL.md— SPF/DKIM/DMARC, React Email, deliverabilityCONTRACT-PDF-GENERATION.md— Puppeteer, react-pdf, CSS Paged MediaCONTRACT-I18N.md— next-intl, ICU, hreflang, RTL, Intl APICONTRACT-ANALYTICS.md— HEART framework, AARRR, GA4, PostHogCONTRACT-PRICING.md— Van Westendorp, EVC, SaaS metrics, tiersCONTRACT-GREEN-SOFTWARE.md— Carbon efficiency, SCI (ISO/IEC 21031:2024)CONTRACT-MOTION-DESIGN.md— UI animation, easing, prefers-reduced-motionCONTRACT-VIDEO-PRODUCTION.md— Video pipeline, platform export specsCONTRACT-PYTHON.md— PEP 8, mypy strict, FastAPI patterns
Read each checked contract directly OR ask Claude: "Read docs/contracts/CONTRACT-X.md before working on [domain]."
Absolute rules
What the AI must NEVER do on this project, specific to this context.
Never do
- [TO FILL IN — e.g.: Modify the Supabase schema without checking RLS]
- [TO FILL IN — e.g.: Switch a page to
'use client'without justification] - [TO FILL IN]
Always do
- [TO FILL IN — e.g.: Run
npm run buildbefore any commit] - [TO FILL IN — e.g.: Check SSR rendering on SEO-critical pages]
- [TO FILL IN]
Error history
Track detected errors here to prevent recurrence. Format: date | error | status
| Date | Error | Status |
| [DD/MM/YYYY] | [description] | Fixed / To fix |
Active ADRs
Architecture Decision Records — the key technical decisions structuring this project. Full template:
[KIT_PATH]/frameworks/ADR-TEMPLATE.mdADR files:./docs/adr/
| ADR | Decision | Status |
| ADR-001 | [TO FILL IN — e.g.: Supabase choice] | Accepted |
| ADR-002 | [TO FILL IN — e.g.: Vercel vs Railway choice] | Accepted |
Create an ADR in ./docs/adr/ for every significant new architectural decision.
SLO (Service Level Objectives)
Target reliability definition. Template:
[KIT_PATH]/frameworks/SLO-TEMPLATE.md
| SLI | Target SLO | Current | Tool |
| Uptime | [TO FILL IN — e.g.: 99.5%] | [current] | UptimeRobot |
| LCP p75 | ≤2.5s | [current] | Vercel Speed Insights |
| Error rate | ≤0.5% | [current] | Sentry |
Error budget: [% SLO] × 30d = [X]h of allowed downtime per month
Tech Debt Tracker
Known technical debt, to be addressed by priority.
| Debt | Impact | Priority | Date detected |
| [TO FILL IN] | [e.g.: Performance] | P1/P2/P3 | [DD/MM] |
Active hooks
Configured in
.claude/settings.json. Review and uncomment to activate.
| Hook | Contract | Trigger | Action |
hook-no-secrets.sh | CONTRACT-SECURITY §2 (OWASP A02) | Pre-commit | BLOCK secrets in staged files |
hook-build-before-commit.sh | AUDIT-DEPLOYMENT | Pre-commit | BLOCK if build/lint fails |
hook-no-dangerous-html.sh | CONTRACT-SECURITY (XSS) | Write *.tsx | WARN on unsafe HTML |
hook-contract-compliance.sh | All active contracts | Post-write | WARN on violations |
hook-audit-before-push.sh | AUDIT-INDEX | Pre-push | REMIND to run audit |
Escape hatch for build hook: SQWR_SKIP_BUILD_CHECK=1 git commit -m "..."
Audit schedule
Run the relevant audit agents before key milestones.
| Trigger | Agents to run | Blocking threshold |
| Before any deployment | AGENT-SECURITY-AUDIT + AGENT-DEPLOYMENT-GATE | Security <70 = BLOCK |
| Before merging a feature | AGENT-CODE-QUALITY-AUDIT + AGENT-SECURITY-AUDIT | Code-Quality <75 |
| Before EU client delivery | + AGENT-ACCESSIBILITY-AUDIT | Accessibility <80 = legal |
| Design / visual change | AGENT-DESIGN-AUDIT + AGENT-PERFORMANCE-AUDIT | Design <70 |
| Adding AI features | AGENT-AI-GOVERNANCE-AUDIT | AI-Gov <80 |
| Production incident | AGENT-OBSERVABILITY-AUDIT | Observability <60 |
| Weekly / end of sprint | AGENT-FULL-AUDIT (all 8 domains) | Global <85 |
Run: ask Claude "Run .claude/agents/AGENT-FULL-AUDIT.md" or use a specific agent.
Mark audit complete: touch .sqwr-last-audit (tracked by hook-audit-before-push)
Quality gates
Workflow for any new feature. Full details:
docs/workflows/WORKFLOW-NEW-FEATURE.md
| Gate | Purpose | Observable truth |
| 0 — Scope | Shape Up Pitch | Scoping doc with Problem + Appetite + No-Gos |
| 1 — Research | Identify applicable contracts | CLAUDE.md "Active contracts" updated |
| 2 — Implement | Code with active hooks | npm run build && npm test exits 0 |
| 3 — Verify | Run audit agents | All scores above thresholds |
| 4 — Ship | CHANGELOG + PR | PR with audit scores, CI passes |
Context notes
[TO FILL IN — important project information that the AI should always keep in mind]
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
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.

