Your AI DevelopmentCommand Center

Comprehensive sync command for any JB Cloud project. Audits and updates every spec, document, UI description, changelog, landing page, and plan file so the entire system reflects the current state of the codebase.

Incrementally fix TypeScript and build errors:

Generate user-friendly changelogs from git commit history

Configure changelog generator settings (categories, voice, output format)

Create or verify a checkpoint in your workflow.

List all available CLI capabilities, MCP servers, and tools Claude can use without asking

Set up a `go.urlstogo.cloud/{repo}--preview` shortlink that auto-updates on every non-main branch push. Designed for Cloudflare Workers projects (not CF Pages or Vercel).

Comprehensive security and quality review of uncommitted changes:

Fetch latest claude-codex configuration from GitHub (browser-safe)

Install claude-codex with auto-sync on a new machine

Bidirectional sync of claude-codex with GitHub from any directory

Export context for claude.ai. Copy output and paste into claude.ai to continue your session there.

Import context from claude.ai back to CLI. Paste the context summary claude.ai gave you.

Restore a previously saved session context. Resume exactly where you left off with full context.

Save current session context for later restoration. Better than /end when you need to pause and resume exactly where you left off.

Create and publish a new release with automatic tag creation and GitHub Release generation

Pre-deployment verification checklist. Run before deploying to catch common issues.

Automatically deploy environment variables from 1Password to a platform.

Audit project dependencies for security vulnerabilities, outdated packages, and unused dependencies.

Generate and run end-to-end tests with Playwright. Creates test journeys, runs tests, captures screenshots/videos/traces, and uploads artifacts.

End session cleanly - sync docs, commit to memory, shutdown. Run when finishing work on a project.

Generate or update `.env.example` files by combining 1Password stored keys with environment variable references found in code.

Manage eval-driven development workflow.

Analyze and fix a GitHub issue by number. Fetches issue details, creates implementation plan, and fixes the code.

Manage project ideas stored in `~/.claude/ideas.md`.

Sync project documentation to jb-cloud-docs (docs.jbcloud.app). Creates or updates project docs in the documentation site.

Analyze the current session and extract any patterns worth saving as skills.

Initialize a new project with full setup - platform selection, style guide, architecture, and implementation plan. Step-by-step with review between phases.

Sequential agent workflow for complex tasks.

Pause session with auto-save and cleanup, resume later with /resume

Restate requirements, assess risks, and create step-by-step implementation plan. WAIT for user CONFIRM before touching any code.

Convert a React/Tailwind component to prompt-kit (ibelick) style.

Safely identify and remove dead code with test verification:

Register a new command or skill convention in the Standards DB. Formats the Invoke field with colon notation and pipe separators, saves to Notion, and updates local skill files.

Remind user what project they're working on and current status

Resume last paused session automatically (no ID needed)

Find successful open-source implementations of a feature, check their licenses for commercial safety, study their patterns, and produce an actionable implementation guide.

Configure routine settings, presets, and behavior

Automated PR merge routine with multi-level review gates (Gemini, security, GitHub Actions)

View routine execution metrics, trends, and optimization suggestions

Smart branch sync with automatic conflict detection and resolution

Save structured context to the Knowledge Base database in Notion. Supports 7 content types with auto-detection, auto-populated git fields, and local backup.

Comprehensive security scan for hardcoded secrets, exposed .env files, and missing 1Password entries.

Comprehensive security audit suite. Run all security checks or specific audits. Use before deployments, after adding auth/API code, or on schedule.

API endpoint security audit. Checks authentication, authorization, input validation, and rate limiting on all routes.

Validate environment variable security. Ensures secrets aren't hardcoded and .env files are properly configured.

Check HTTP security headers against OWASP recommendations. Validates CSP, HSTS, and other protective headers.

Check dependency licenses for compliance. Flags GPL/AGPL in commercial projects and identifies unknown licenses.

Deep secrets detection in code and git history. Finds API keys, tokens, passwords, and other sensitive data.

Check branch and context at session start. Run this first when resuming work to avoid branch confusion.

Set up GitHub Actions workflows with intelligent defaults based on project type

Configure your preferred package manager (npm/pnpm/yarn/bun)

Configure a stable preview subdomain on Vercel with automatic branch sync and search engine blocking.

Generate standup notes from recent git activity and session context. Perfect for daily standups or status updates.

Begin session - check branch, load context, verify ready state. Run this first when starting work.

Sync the plans index at ~/Desktop/plans-index.html with the current state of ~/.claude/plans/.

Enforce test-driven development workflow. Scaffold interfaces, generate tests FIRST, then implement minimal code to pass. Ensure 80%+ coverage.

Analyze test coverage and generate missing tests:

Analyze the codebase structure and update architecture documentation:

Sync documentation from source-of-truth:

Run comprehensive verification on current codebase state.

Multi-provider AI integration specialist. Invoke when setting up AI providers in a new or existing Next.js project. Handles both the settings panel (user key management) and the runtime side (provider resolution, agent loop, chat API). Use after /add-ai-providers for the runtime wiring, or standalon

Software architecture specialist for system design, scalability, and technical decision-making. Use PROACTIVELY when planning new features, refactoring large systems, or making architectural decisions.

Expert Bricks Builder and AutomaticCSS developer for generating valid Bricks JSON from screenshots

Build and TypeScript error resolution specialist. Use PROACTIVELY when build fails or type errors occur. Fixes build/type errors only with minimal diffs, no architectural edits. Focuses on getting the build green quickly.

Expert code review specialist. Proactively reviews code for quality, security, and maintainability. Use immediately after writing or modifying code. MUST BE USED for all code changes.

Analyze conversation transcripts to find behaviors worth preventing with hooks.

Documentation and codemap specialist. Use PROACTIVELY for updating codemaps and documentation. Runs /update-codemaps and /update-docs, generates docs/CODEMAPS/*, updates READMEs and guides.

End-to-end testing specialist using Playwright. Use PROACTIVELY for generating, maintaining, and running E2E tests. Manages test journeys, quarantines flaky tests, uploads artifacts (screenshots, videos, traces), and ensures critical user flows work.

Environment variable deployment specialist for detecting and deploying env vars to platforms

Converts React components to ibelick's style — prompt-kit and ai-elements quality. Use when the user wants to refactor a component, convert a shadcn component, or build a new component in this style. Produces compound components with React Context, clean minimal JSX, AI-aware status machines, and ib

Documentation sync agent. Automatically documents projects to jb-cloud-docs repository (docs.jbcloud.app). Creates/updates project documentation with architecture, plans, and progress.

Expert planning specialist for complex features and refactoring. Use PROACTIVELY when users request feature implementation, architectural changes, or complex refactoring. Automatically activated for planning tasks.

Dead code cleanup and consolidation specialist. Use PROACTIVELY for removing unused code, duplicates, and refactoring. Runs analysis tools (knip, depcheck, ts-prune) to identify dead code and safely removes it.

Security vulnerability detection and remediation specialist. Use PROACTIVELY after writing code that handles user input, authentication, API endpoints, or sensitive data. Flags secrets, SSRF, injection, unsafe crypto, and OWASP Top 10 vulnerabilities.

Test-Driven Development specialist enforcing write-tests-first methodology. Use PROACTIVELY when writing new features, fixing bugs, or refactoring code. Ensures 80%+ test coverage.

Security reviewer for wp-dispatch crypto, auth, and SSH-adjacent code. Use when editing crypto.ts, auth.ts, proxy.ts, or any endpoint handling credentials. Reviews for timing attacks, key material exposure, JWT misconfiguration, and auth bypass vectors.

Add multi-provider AI settings panel to a JB Cloud Next.js app. Copies AIProvidersPanel component from ~/clarity, creates API routes per provider, adds integrations table to schema, and wires up the settings page. Use when an app needs user-managed AI API keys with encrypted storage and automatic fa

Backend architecture patterns, API design, database optimization, and server-side best practices for Node.js, Express, and Next.js API routes.

**Command:** `/bricks`

**Command:** `/bricks-learn`

Three-tier changelog standard for all JB Cloud projects. Contains verbatim templates — use this before writing any changelog entry.

Toggle dev changelog tracking. Redirects to the three-tier /changelog standard.

Automatically creates user-facing changelogs from git commits by analyzing commit history, categorizing changes, and transforming technical commits into clear, customer-friendly release notes.

Analyze a codebase and recommend Claude Code automations (hooks, subagents, skills, plugins, MCP servers). Use when setting up Claude Code for a project or optimizing workflows.

Audit and improve CLAUDE.md files in repositories. Scans for all CLAUDE.md files, evaluates quality against criteria, outputs quality report, then makes targeted updates.

ClickHouse database patterns, query optimization, analytics, and data engineering best practices for high-performance analytical workloads.

Universal coding standards, best practices, and patterns for TypeScript, JavaScript, React, and Node.js development.

Automatically extract reusable patterns from Claude Code sessions and save them as learned skills for future use.

Analyze any dashboard page and generate an interactive HTML mockup with horizontal tabs. Use when reorganizing flat/scrolling dashboard pages into tabbed layouts.

A formal evaluation framework for Claude Code sessions, implementing eval-driven development (EDD) principles.

Frontend development patterns for React, Next.js, state management, performance optimization, and UI best practices.

Automate GitHub repositories, issues, pull requests, branches, CI/CD, and permissions via gh CLI. Manage code workflows, review PRs, search code, and handle deployments programmatically.

Use when adding an in-app help/onboarding hub to a dashboard app. Triggers when a project needs a Getting Started page, feature guide, or in-app documentation with tabbed navigation. Also use when the existing help is buried in Settings and needs to be promoted to a first-class nav item.

Use when syncing a project's help hub implementation with the latest help-hub skill standards. Audits the help page for incorrect card spacing, missing deep links to settings panels, and outdated Connections Guide links. Run after updating the help-hub skill or when a project's help page was built b

DEPRECATED: Use /hig-ui-pass instead. hig-mobile-pass has been superseded by the full-platform hig-ui plugin which covers Mobile, iPad, Desktop, CSS token audit, PWA asset generation, and web standards.

Full Apple HIG + web standards compliance pass covering Mobile, iPad, and Desktop. Replaces hig-mobile-pass. Covers CSS token audit, platform-specific checklists, PWA asset generation, web standards (Vercel WIG), and design consistency audit.

Sync all HIPAA documentation — update memory files, regenerate PDFs from HTML source files, regenerate DOCX, archive old interaction log entries, and verify cross-document consistency. Run at the end of any HIPAA-related session.

Guide for creating hookify rules - markdown files with YAML frontmatter that define patterns to watch for and messages to show when matched.

- Type: Electron desktop app

**Extracted:** 2026-01-29

**Extracted:** 2026-02-23

**Extracted:** 2025-02-01

**Last Session:** 2026-01-31

**Extracted:** 2026-02-19

**Extracted:** 2026-02-12

**Extracted:** 2026-02-12

**Extracted:** 2026-02-12

**Extracted:** 2026-02-18

**Extracted:** 2026-02-14

**Extracted:** 2026-02-18

In Astro hybrid mode (`output: 'hybrid'`), dynamic routes like `[id].astro` or `[slug].astro` fail at build time with:

**Extracted:** 2026-02-18

When using React context hooks (e.g., `useToast`, `useAuth`) inside Astro components rendered with `client:load`, the context provider may not wrap the component during SSR. Astro renders each `client:load` island independently on the server, so a provider in a parent island doesn't wrap child islan

`workarounds` | `project_specific`

**Extracted:** 2026-02-19

**Extracted:** 2026-02-19

**Extracted:** 2026-01-25

**Extracted:** 2026-01-28

**Extracted:** 2026-01-28

- Type: Web app (Teaching System MVP)

**Extracted:** January 2025

**Extracted:** 2026-01-25

**Extracted:** 2026-02-13

**Extracted:** 2026-02-08

**Extracted:** 2026-02-17

**Extracted:** 2026-02-18

**Extracted:** 2026-02-15

**Extracted:** 2026-02-18

**Extracted:** 2026-02-19

Two Chrome MCP servers exist (`chrome-devtools` and `claude-in-chrome`) with separate tab/page contexts. Screenshots and actions can target the wrong tab if using the wrong server.

**Extracted:** 2026-02-23

**Project:** claude-command (MC3)

**Extracted:** 2026-02-20

**Extracted:** 2026-02-05

**Category**: Backend Patterns

**Extracted:** 2026-02-19

**Extracted:** 2026-02-19

**Extracted:** 2026-02-18

**Extracted:** 2026-02-21

You have an existing display component (e.g., StatCard) and want to make some instances clickable/navigable without wrapping each usage in a `<Link>` at the call site.

**Extracted:** 2026-02-19

**Extracted:** 2026-02-17

**Extracted:** 2026-02-17

**Extracted:** 2026-01-31

Implementing drag-and-drop list reordering in React with @dnd-kit requires careful handling of sortable items, drag handles, and state mutations.

**Extracted:** 2026-02-16

The `Edit` tool's `replace_all: true` flag matches strings exactly, including indentation. The same semantic pattern at different indentation levels will NOT be matched by a single `replace_all` call.

**Extracted:** 2026-02-14

- Type: Chrome Extension + MCP Server

**Extracted:** 2026-02-18

**Extracted:** 2026-01-25

**Extracted:** 2026-01-25

`if (secret) { check }` is fail-open — if env var is unset, auth is bypassed entirely. Fix: fail closed with 503 when secret is missing, then 401 if wrong. Applies to cron endpoints, webhooks, any bearer-token auth.

- Type: Web application

**Last Updated:** January 29, 2026

- Type: Documentation site

- Created: 2025-01-27

- Created: 2026-01-28

**Extracted:** 2026-02-21

Building interactive mini-component previews for a marketplace/catalog UI where each card shows a functional demo of the component it represents.

`error_resolution` | `debugging_techniques`

**Extracted:** 2026-02-19

**Status:** Idea - needs debugging

**Extracted:** 2026-02-18

- **Type:** Dashboard application

**Extracted:** 2026-01-28

**Extracted:** 2026-02-15

MEMORY.md auto-loads into every Claude Code session — but only the **first 200 lines**. Anything past line 200 is silently cut off.

**Extracted:** 2026-02-21

Runtime patterns for multi-provider AI chat using Vercel AI SDK. Covers provider resolution, SDK wiring gotchas, agent loop with tool calling, summary generation, and error classification. Discovered building WP Dispatch builds 14.9-14.15.

**Extracted:** 2026-02-23

**Extracted:** 2026-01-29

**Extracted:** 2026-02-22

**Extracted:** 2026-02-22

Next.js App Router route groups use parentheses in directory names (e.g., `(dashboard)`), which require special handling in shell commands. Bare paths fail; proper quoting is essential.

**Extracted:** 2026-01-29

**Extracted:** 2026-02-19

**Extracted:** 2026-02-16

**Extracted:** 2026-02-21

**Extracted:** 2026-02-20

The Notion MCP `notion-create-pages` tool rejects multi_select field values that are passed as comma-separated strings.

**Extracted:** 2026-02-15

**Extracted:** 2026-02-23

**Extracted:** 2026-02-13

**Extracted:** 2026-02-21

**Extracted:** 2026-02-12

**Extracted:** 2026-02-05

**Extracted:** 2026-02-19

**Extracted:** 2026-02-16

**Extracted:** 2026-02-20

- **Type**: Healthcare EMR Companion App

- Type: macOS menu bar utility app

**Category**: Development Workflow

**Extracted:** 2026-02-18

- Type: Command System / Automation Workflow

- Type: CLI automation framework

**Extracted:** 2026-02-22

`error_resolution` | `debugging_techniques`

**Date:** 2026-02-01

- jb-cloud-docs (Aventerica89/jb-cloud-docs)

After squash-merging PRs, branches become orphaned because:

**Extracted:** 2026-02-05

**Extracted:** 2026-02-22

**Extracted:** 2026-01-29

When updating shadcn/ui component interfaces (adding new props or changing prop types), the registry copies under `registry/new-york/` must be kept in sync, or build fails with type errors.

**Extracted:** 2026-02-05

**Extracted:** 2026-02-18

When you want Claude to behave consistently across different contexts — CLI sessions, VaporForge sandboxes, fresh sessions with no memory — put the standard in a **skill**, not a Notion database or MEMORY.md.

**Extracted:** 2026-02-22

**Extracted:** 2026-02-15

Git history is not available at runtime on serverless platforms (Vercel, Cloudflare Workers). You can't dynamically generate changelogs from `git log`.

**Extracted:** 2026-02-08

- **Session Date**: 2026-01-28

**Extracted:** 2026-02-21

The `supabase db push` command fails when the local `supabase/migrations/` directory contains multiple files with the same numeric prefix (e.g., `005_create_tables.sql` and `005_app_todos.sql`). The CLI tries to apply both and fails if one is already on the remote.

**Extracted:** 2026-02-18

**Extracted:** 2026-02-21

**Extracted:** 2026-02-21

**Extracted:** 2026-02-18

**Extracted:** 2026-02-22

**Extracted:** 2026-02-18

**Extracted:** 2026-02-19

**Extracted:** 2026-02-22

**Extracted:** 2026-02-21

**Extracted:** 2026-02-22

**Extracted:** 2026-01-31

- Type: URL shortener (Cloudflare Worker)

**Extracted:** 2026-02-14

**Extracted:** 2026-02-05

- Adding a new AI provider (OpenAI, Mistral, etc.) to VaporForge

**Extracted:** 2026-01-29

**Extracted:** 2026-02-08

When a Vercel deployment fails after `git push` and you need to diagnose and fix the build error without leaving the CLI.

**Extracted:** 2026-01-29

**Extracted:** 2026-02-23

**Extracted:** 2026-02-21

**Extracted:** 2026-02-19

- Type: Web App (Dashboard)

- Type: WordPress site management dashboard

**Extracted:** 2026-02-16

**Extracted:** 2026-02-19

**Extracted:** 2026-02-18

- Adding a new AI provider or external API as an MCP tool in a sandboxed environment

**Extracted:** 2026-01-29

Master iOS Human Interface Guidelines and SwiftUI patterns for building native iOS apps. Use when designing iOS interfaces, implementing SwiftUI views, or ensuring apps follow Apple's design principles.

Creates interactive HTML playgrounds - self-contained single-file explorers with controls, live preview, and prompt output. Use when user asks for a playground, explorer, or interactive tool.

This is an example of a project-specific skill. Use this as a template for your own projects.

Activates when the user asks about AI prompts, needs prompt templates, wants to search for prompts, or mentions prompts.chat. Use for discovering, retrieving, and improving prompts.

You are running a smart PR merge routine with configurable review gates.

You are running a smart PR merge routine with automatic session cleanup.

Save an architectural decision record (ADR) to Notion Knowledge Base. Shortcut for /save-to-notion with type=decision. Use after choosing between approaches.

Save an implementation plan to Notion Knowledge Base. Shortcut for /save-to-notion with type=plan. Use when completing or pausing work on a multi-phase plan.

Register a command or workflow convention in the Notion Standards DB. Shortcut for /save-to-notion standards. Use after finalizing a new command pattern or system convention.

Save session context, plans, decisions, or patterns to the local Obsidian vault. Zero token overhead - writes pure markdown directly to the file system. Use instead of or alongside save-to-notion.

Use this skill when adding authentication, handling user input, working with secrets, creating API endpoints, or implementing payment/sensitive features. Provides comprehensive security checklist and patterns.

Set up Vercel preview subdomain with auto-sync branch and SEO protection. Creates {repo-name}-preview.jbcloud.app pattern.

Suggests manual context compaction at logical intervals to preserve context through task phases rather than arbitrary auto-compaction.

Use this skill when writing new features, fixing bugs, or refactoring code. Enforces test-driven development with 80%+ coverage including unit, integration, and E2E tests.

A comprehensive verification system for Claude Code sessions.

Located in `~/.claude/agents/`:

When auto-remind is enabled, Claude should periodically show a brief context reminder to help users track what they're working on.

**ALWAYS check available tools and capabilities BEFORE asking the user for information.**

ALWAYS create new objects, NEVER mutate:

PROACTIVELY offer to deploy env vars when you detect:

```

- **PreToolUse**: Before tool execution (validation, parameter modification)

**ALWAYS save session context to Notion before compacting.** Context lost during compaction is gone forever.

```typescript

**Haiku 4.5** (90% of Sonnet capability, 3x cost savings):

Before ANY commit:

At the START of every session, BEFORE doing any work, run the `/start` command or follow this checklist:

Long inline strings cause:

Test Types (ALL required):