v2.1.140 · Claude Opus 4.6 · 1M context

Claude Code Feature Showcase

A fully loaded AI development environment with plugins, skills, specialized agents, and deep integrations — all configured for high-effort, autonomous engineering workflows.

6Plugins
50+Skills
50+Agents
20+Built-in Tools
4API Integrations
12Projects
Installed Plugins

Superpowers v5.0.5

Enhanced workflow skills: brainstorming, planning, TDD, debugging, code review, git worktrees, parallel agent dispatch, and verification before completion.

claude-plugins-official

Compound Engineering v2.59.0

Full-stack engineering toolkit: frontend design, browser testing, document review, git workflows, research agents, 20+ code review personas.

EveryInc

Context7

Live documentation fetcher for any library, framework, or SDK. Queries current docs even when training data is outdated.

claude-plugins-official

Code Review

PR review skill that orchestrates multi-persona code review with tiered agents for correctness, security, performance, and style.

claude-plugins-official

GitHub

Native GitHub integration for issues, pull requests, checks, releases, and repository management via the gh CLI.

claude-plugins-official

Swift LSP

Language Server Protocol support for Swift development with code intelligence, diagnostics, and completions.

claude-plugins-official
Development Workflow Pattern

The full skill-driven lifecycle. Each stage is backed by a dedicated skill that enforces engineering rigor.

Step 1
Brainstorm
/brainstorming
Step 2
Write Plan
/writing-plans
Step 3
Execute
/executing-plans
Step 4
TDD
/test-driven-dev
Step 5
Review
/ce-review
Step 6
Verify
/verification
Step 7
Ship
/finishing-branch

Parallel Execution

Independent tasks are dispatched to parallel sub-agents via /dispatching-parallel-agents or /subagent-driven-development. Each agent works in isolation, then results merge back.

Git Worktrees

Feature work is isolated using /using-git-worktrees. Each worktree gets its own branch and working directory so parallel features never conflict.

Systematic Debugging

Bugs trigger /systematic-debugging before any fix is proposed. Root cause analysis first, then targeted fix, then verification.

Persistent Memory

File-based memory stores user preferences, feedback, project context, and external references across sessions. Indexed via MEMORY.md.

Configuration (settings.json)
settings.json
{ "enabledPlugins": { "code-review@claude-plugins-official": true, "context7@claude-plugins-official": true, "github@claude-plugins-official": true, "swift-lsp@claude-plugins-official": true, "superpowers@claude-plugins-official": true, "compound-engineering@compound-engineering-plugin": true }, "extraKnownMarketplaces": { "compound-engineering-plugin": { "source": { "source": "github", "repo": "EveryInc/compound-engineering-plugin" } } }, "effortLevel": "high", // max reasoning depth "skipDangerousModePermissionPrompt": true // faster iteration }
Available Skills (Slash Commands)

Superpowers

/brainstorming /writing-plans /executing-plans /test-driven-development /systematic-debugging /requesting-code-review /receiving-code-review /verification-before-completion /finishing-a-development-branch /dispatching-parallel-agents /subagent-driven-development /using-git-worktrees /writing-skills

Compound Engineering — Workflow

/ce-brainstorm /ce-plan /ce-work /ce-review /ce-ideate /ce-compound /git-commit /git-commit-push-pr /git-clean-gone-branches /git-worktree /reproduce-bug /todo-resolve /claude-permissions-optimizer

Compound Engineering — Design & Docs

/frontend-design /agent-browser /test-browser /feature-video /gemini-imagegen /onboarding /proof /document-review /every-style-editor

Framework & Utilities

/claude-api /agent-native-architecture /dhh-rails-style /andrew-kane-gem-writer /dspy-ruby /rclone /simplify /loop /schedule /update-config /keybindings-help /apply-anthropic-skill-best-practices
Specialized Sub-Agents

Autonomous agents dispatched for specific tasks. Can run in parallel or foreground, each with isolated context.

Core

Explore

Fast codebase exploration — find files, search code, answer architecture questions.

Plan

Software architect for designing implementation strategies and identifying trade-offs.

claude-code-guide

Answers questions about Claude Code features, hooks, MCP servers, and SDK usage.

Research

framework-docs-researcher

Gathers official docs and best practices for any framework or library.

git-history-analyzer

Archaeological git analysis — trace code evolution, understand why patterns exist.

repo-research-analyst

Deep repo structure, conventions, patterns, and documentation analysis.

issue-intelligence-analyst

GitHub issue analysis — recurring themes, pain patterns, severity trends.

best-practices-researcher

Researches external best practices, community conventions, and implementation guidance.

learnings-researcher

Searches past solutions to surface institutional knowledge and prevent repeated mistakes.

Code Review (27 personas)

correctness-reviewer

Always-on. Logic errors, edge cases, state management bugs, intent mismatches.

maintainability-reviewer

Always-on. Premature abstraction, dead code, coupling, naming clarity.

testing-reviewer

Always-on. Coverage gaps, weak assertions, brittle tests, missing edge cases.

project-standards-reviewer

Always-on. Audits changes against CLAUDE.md/AGENTS.md standards.

security-reviewer

Auth, endpoints, user input, permissions — exploitable vulnerabilities.

performance-reviewer

DB queries, loops, caching, I/O — scalability and runtime issues.

adversarial-reviewer

Constructs failure scenarios for large or high-risk diffs.

api-contract-reviewer

API routes, serialization, versioning — breaking contract changes.

data-migrations-reviewer

Migration files, schema changes, data integrity, backfill safety.

reliability-reviewer

Error handling, retries, circuit breakers, timeouts, health checks.

architecture-strategist

Pattern compliance, design integrity, structural refactors.

code-simplicity-reviewer

Final pass for YAGNI violations and simplification opportunities.

security-sentinel

OWASP compliance, input validation, hardcoded secrets, auth audits.

performance-oracle

Algorithmic complexity, memory usage, DB queries, scalability.

agent-native-reviewer

Ensures agent-native parity — any user action, an agent can also take.

pattern-recognition-specialist

Design patterns, anti-patterns, naming conventions, duplication.

data-integrity-guardian

Migration safety, data constraints, transaction boundaries, privacy.

schema-drift-detector

Cross-references schema.rb changes against included migrations.

kieran-typescript-reviewer

Strict bar for type safety, clarity, and maintainability in TypeScript.

kieran-rails-reviewer

Strict bar for clarity, conventions, and maintainability in Rails.

kieran-python-reviewer

Pythonic clarity, type hints, and maintainability.

dhh-rails-reviewer

Opinionated DHH perspective on Rails architecture and patterns.

previous-comments-reviewer

Checks whether prior PR feedback has been addressed in the current diff.

deployment-verification-agent

Go/No-Go checklists, SQL verification, rollback procedures.

cli-agent-readiness-reviewer

Reviews CLIs for AI agent readiness using severity-based rubric.

Design & Documents

design-iterator

Iterative UI refinement through N screenshot-analyze-improve cycles.

design-implementation-reviewer

Compares live UI against Figma designs for design fidelity feedback.

figma-design-sync

Detects and fixes visual differences between implementation and Figma.

adversarial-document-reviewer

Challenges premises, surfaces assumptions, stress-tests decisions.

coherence-reviewer

Reviews for internal consistency, contradictions, and terminology drift.

product-lens-reviewer

Senior product leader perspective on scope and goal alignment.

design-lens-reviewer

Missing design decisions: IA, interaction states, user flows.

security-lens-reviewer

Plan-level security gaps: auth assumptions, data exposure, threat models.

feasibility-reviewer

Evaluates if proposed approaches survive reality.

scope-guardian-reviewer

Challenges unnecessary abstractions and scope creep.

spec-flow-analyzer

Specs for flow completeness, edge cases, and requirement gaps.

Workflow

bug-reproduction-validator

Systematically reproduces and validates bug reports before fixing.

lint

Runs linting and code quality checks on Ruby and ERB files.

pr-comment-resolver

Evaluates and resolves PR review threads with structured summaries.

ankane-readme-writer

Creates READMEs following Ankane-style template for Ruby gems.

Built-in Tools
File Operations
ReadWriteEditGlobGrepNotebookEdit
Execution
BashLSP
Web & Search
WebFetchWebSearchToolSearch
Task Management
TaskCreateTaskUpdateTaskListTaskGetTaskStopTaskOutput
Planning & Orchestration
AgentSkillEnterPlanModeExitPlanModeEnterWorktreeExitWorktree
Scheduling & Remote
CronCreateCronDeleteCronListRemoteTriggerAskUserQuestion
API Integrations (MCP Servers)

Context7

Live documentation queries for any library, framework, or SDK. Two instances running (plugin + standalone) for redundancy.

Gmail

Read, search, and manage email directly from the coding environment via OAuth authentication.

Google Calendar

View and manage calendar events, check availability, and create meetings.

Notion

Access Notion workspaces — read pages, query databases, manage content.

Pre-Approved Permissions

Commands auto-allowed without confirmation prompt, configured in settings.local.json.

python / python3 pip / pip3 install npm install / build / check bun mkdir / chmod tree / ls / find curl / ping timeout kill / pkill open pdflatex / pandoc huggingface-cli WebSearch WebFetch (cdnjs, metorial, stack-ai)
Key Design Decisions

High Effort, Always

effortLevel is set to "high" globally. Every task gets maximum reasoning depth — no shortcuts, no shallow passes.

Skill-First Workflow

Before any action, relevant skills are checked and invoked. Skills enforce discipline: brainstorm before build, verify before ship.

Dangerous Mode Enabled

Permission prompts are skipped for faster iteration. Pre-approved commands cover the common toolchain (Python, Node, git, etc.).

Multi-Marketplace Plugins

Draws from 3 plugin sources: Anthropic Official, EveryInc (Compound Engineering), and thedotmack (claude-mem). Community + first-party.

Agent-Native Architecture

30+ specialized sub-agents for research, review, design, and workflow. Tasks are parallelized and isolated by default.

Cross-Session Memory

Persistent file-based memory indexed by MEMORY.md. Stores user context, feedback, project state, and external references across conversations.

Personal Usage Dashboard

Lifetime Claude Code activity for swchoi1994. Snapshot as of 2026-04-17.

Less More

You’ve used ~289× more tokens than Animal Farm.