{"id":49416483,"url":"https://github.com/chris2ao/claude-code-config","last_synced_at":"2026-04-29T03:10:46.222Z","repository":{"id":337130861,"uuid":"1152418092","full_name":"chris2ao/claude-code-config","owner":"chris2ao","description":null,"archived":false,"fork":false,"pushed_at":"2026-04-25T03:37:38.000Z","size":842,"stargazers_count":0,"open_issues_count":0,"forks_count":0,"subscribers_count":0,"default_branch":"master","last_synced_at":"2026-04-25T05:26:33.811Z","etag":null,"topics":["claude-code","configuration","mcp","rules","skills"],"latest_commit_sha":null,"homepage":null,"language":"Shell","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":null,"status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/chris2ao.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":null,"code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":"docs/security-reviews/pixel-agents-review.md","support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null,"zenodo":null,"notice":null,"maintainers":null,"copyright":null,"agents":null,"dco":null,"cla":null}},"created_at":"2026-02-07T21:04:55.000Z","updated_at":"2026-04-25T03:37:42.000Z","dependencies_parsed_at":null,"dependency_job_id":null,"html_url":"https://github.com/chris2ao/claude-code-config","commit_stats":null,"previous_names":["chris2ao/claude-code-config"],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/chris2ao/claude-code-config","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/chris2ao%2Fclaude-code-config","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/chris2ao%2Fclaude-code-config/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/chris2ao%2Fclaude-code-config/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/chris2ao%2Fclaude-code-config/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/chris2ao","download_url":"https://codeload.github.com/chris2ao/claude-code-config/tar.gz/refs/heads/master","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/chris2ao%2Fclaude-code-config/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":32408549,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-04-29T02:37:21.628Z","status":"ssl_error","status_checked_at":"2026-04-29T02:36:50.947Z","response_time":110,"last_error":"SSL_connect returned=1 errno=0 peeraddr=140.82.121.5:443 state=error: unexpected eof while reading","robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":false,"can_crawl_api":true,"host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["claude-code","configuration","mcp","rules","skills"],"created_at":"2026-04-29T03:10:45.525Z","updated_at":"2026-04-29T03:10:46.202Z","avatar_url":"https://github.com/chris2ao.png","language":"Shell","funding_links":[],"categories":[],"sub_categories":[],"readme":"# Claude Code Configuration\n\nA production-ready configuration for [Claude Code](https://docs.claude.com/en/docs/claude-code) with 15 rules, 34 agents, 28 invocable skills, 23 learned skills, 27 scripts, 11 commands, 12 hooks, 7 MCP servers, and 50 instincts. Built through months of daily use across multiple projects on macOS and Windows.\n\n## What This Is\n\nThis repo contains a complete, portable Claude Code setup that turns a single AI assistant into a team of specialized agents with persistent memory, automated hooks, and hard-won debugging knowledge. Copy what you need, adapt it to your workflow, and skip the trial-and-error we already went through.\n\nFor a detailed walkthrough aimed at beginners, see [COMPLETE-GUIDE.md](./COMPLETE-GUIDE.md).\n\n## Quick Start\n\n```bash\n# 1. Copy rules (works standalone, no dependencies)\ncp -r rules/ ~/.claude/rules/\n\n# 2. Add MCP servers (requires Node.js)\nclaude mcp add --scope user memory -- npx -y @modelcontextprotocol/server-memory\nclaude mcp add --scope user context7 -- npx -y @upstash/context7-mcp@latest\n\n# 3. Copy custom agents\ncp -r agents/ ~/.claude/agents/\n\n# 4. Install hooks (copy scripts, then configure)\ncp -r hooks/ ~/.claude/hooks/\n# Edit hooks/settings.local.json.template and copy to your project's .claude/settings.local.json\n\n# 5. Copy skills\ncp -r skills/ ~/.claude/skills/\n\n# 6. (Optional) Install the everything-claude-code plugin for plugin agents\n# Inside a Claude Code session:\n/plugin marketplace add affaan-m/everything-claude-code\n/plugin install everything-claude-code@everything-claude-code\n```\n\n## Component Inventory\n\n### Rules (15 files)\n\nRules in `rules/` are loaded automatically into every Claude Code session. They shape how Claude writes code, handles security, manages git, and routes work to agents.\n\n| File | Purpose |\n|------|---------|\n| `agents.md` | Agent orchestration framework (plugin + custom agents) |\n| `core/agentic-workflow.md` | Mandatory parallel task decomposition with model routing |\n| `core/coding-style.md` | Immutability, file organization, error handling, no em dashes |\n| `core/memory-management.md` | Five memory systems with clear boundaries and save triggers |\n| `core/security.md` | Pre-commit security checklist, secret management protocol |\n| `development/git-workflow.md` | Conventional commits, PR workflow, feature implementation |\n| `development/patterns.md` | Repository pattern, API response envelopes, skeleton projects |\n| `development/testing.md` | TDD workflow, 80% minimum coverage |\n| `operations/hooks.md` | Hook types, file protection, context preservation |\n| `operations/performance.md` | Model selection (Haiku/Sonnet/Opus), cost optimization |\n| `operations/windows-platform.md` | PowerShell stdin, Git Bash path mangling, OneDrive locks |\n| `content/blog-content.md` | Blog writing rules (no private repo links) |\n| `operations/context-preservation.md` | Session context preservation across compactions |\n| `operations/macos-platform.md` | macOS shell, Homebrew, notifications, file system |\n\n### Agents (34 files)\n\nAgents in `agents/` are specialized agent definitions spawned via Claude Code's Task tool. Each has a focused role and optimal model assignment.\n\n**Core Agents** (18):\n\n| Agent | Model | Purpose |\n|-------|-------|---------|\n| `changelog-writer` | haiku | Auto-generate CHANGELOG.md entries from git diffs and session context |\n| `config-sync` | haiku | Compare local Claude Code config against claude-code-config repo |\n| `context-health` | haiku | Monitor context window usage and suggest compaction points |\n| `deploy-verifier` | haiku | Captain agent: end-to-end deploy verification with parallel checks |\n| `evolve-synthesizer` | sonnet | Synthesizes instinct clusters into evolved agent, skill, and command candidates |\n| `gmail-assistant` | sonnet | Daily Gmail inbox cleanup: content-aware classification, auto-labeling, VIP detection, follow-up tracking |\n| `home-sync` | haiku | Harvest and sync config artifacts from all repos |\n| `multi-repo-orchestrator` | haiku | Captain agent: parallel git operations across all project repos |\n| `notebooklm-assistant` | sonnet | Orchestrates NotebookLM workflows: notebooks, sources, content generation, research, downloads |\n| `notebooklm-content` | sonnet | Creates branded infographics and slide decks from blog posts using Google NotebookLM |\n| `pre-commit-checker` | inherit | Unified pre-commit security and code quality gate |\n| `refine-captain` | opus | /refine orchestrator: evidence-based component refinement from session transcripts |\n| `refine-reader` | haiku | /refine evidence reader: extracts relevant transcript excerpts for component editing |\n| `session-analyzer` | sonnet | Captain agent: parallel session transcript analysis with synthesis |\n| `session-checkpoint` | haiku | Lightweight mid-session state preservation before context compaction |\n| `skill-extractor` | sonnet | Captain agent: parallel instinct extraction from transcripts |\n| `sync-orchestrator` | haiku | Bidirectional config sync with security scanning |\n| `wrap-up-orchestrator` | haiku | Automated session wrap-up for multi-repo workflows |\n\n**Game Development Team** (6):\n\n| Agent | Model | Purpose |\n|-------|-------|---------|\n| `game-artist` | sonnet | Game visual artist: sprites, animations, CSS styling, canvas rendering |\n| `game-designer` | sonnet | Game mechanics designer: core loop, systems, balance, progression |\n| `game-developer` | sonnet | Game developer: engine logic, state management, game loop, physics, AI |\n| `game-director` | opus | Captain agent: orchestrates game development team |\n| `game-ux` | sonnet | Game UX/UI designer: menus, HUD, player feedback, accessibility |\n| `game-writer` | haiku | Game writer: story, dialogue, world-building, lore, tutorial text |\n\n**Blog Production Team** (5):\n\n| Agent | Model | Purpose |\n|-------|-------|---------|\n| `blog-captain` | opus | Captain agent: orchestrates multi-agent blog post production pipeline |\n| `blog-editor` | sonnet | Senior blog editor: reviews posts for hooks, pacing, entertainment |\n| `blog-ux` | haiku | Blog UX/UI agent: build verification and structural analysis of MDX |\n| `blog-voice` | sonnet | Blog voice agent: maintains living voice profile, produces voice briefs |\n| `blog-writer` | sonnet | Blog post writer: drafts and revises MDX posts for CryptoFlex LLC |\n\n**UI/UX Team** (5):\n\n| Agent | Model | Purpose |\n|-------|-------|---------|\n| `ui-component-architect` | sonnet | UI/UX Component Architect: design tokens, composition patterns, responsive design, semantic HTML, Tailwind |\n| `ui-performance-reviewer` | haiku | UI/UX Performance Reviewer: bundle size, Core Web Vitals, React/Next.js patterns, server vs client analysis |\n| `ui-ux-director` | sonnet | Captain agent: orchestrates UI/UX design team across design, build, review, and audit workflows |\n| `ui-ux-reviewer` | sonnet | UI/UX Reviewer + QA: heuristics evaluation, TASTE scoring, anti-pattern detection, Playwright visual testing, final quality gate |\n| `ui-visual-designer` | sonnet | UI/UX Visual Designer: aesthetic direction, color systems, typography, layout composition, anti-AI-slop |\n\n### Superpowers Plugin Skills (14 skills + 1 agent)\n\nThe [superpowers plugin](https://github.com/anthropics/claude-plugins-official) provides a structured development workflow. Skills activate automatically based on context. Standard flow: brainstorming -\u003e using-git-worktrees -\u003e writing-plans -\u003e subagent-driven-development -\u003e test-driven-development -\u003e requesting-code-review -\u003e finishing-a-development-branch.\n\n| Skill | Trigger | What It Does |\n|-------|---------|-------------|\n| `using-superpowers` | Auto (session start) | Gateway: explains how to find and invoke all other skills |\n| `brainstorming` | Before creative work | Socratic design refinement, saves spec to `docs/superpowers/specs/` |\n| `writing-plans` | After brainstorming | Breaks designs into 2-5 min tasks with file paths, code, and verification |\n| `executing-plans` | Offline plan execution | Loads plan, executes tasks with verification checkpoints |\n| `subagent-driven-development` | In-session plan execution | One subagent per task with two-stage review |\n| `test-driven-development` | Before production code | RED-GREEN-REFACTOR cycle |\n| `systematic-debugging` | Bug encountered | 4-phase root cause: investigate, analyze, hypothesize, implement |\n| `verification-before-completion` | Before claiming done | Evidence-before-claims verification |\n| `using-git-worktrees` | After design approval | Isolated workspace with .gitignore verification |\n| `finishing-a-development-branch` | After implementation | Merge/PR/keep/discard options, worktree cleanup |\n| `requesting-code-review` | After completing tasks | Dispatches code-reviewer agent |\n| `receiving-code-review` | Receiving feedback | Technical evaluation before implementing changes |\n| `dispatching-parallel-agents` | 2+ independent problems | Concurrent subagents for independent tasks |\n| `writing-skills` | Creating new skills | TDD methodology for skill authoring |\n\nAgent: `code-reviewer` reviews completed work against plans for quality, architecture, and docs.\n\n### Skills (28 invocable + 23 learned)\n\n**Invocable skills** (in `skills/*/SKILL.md`) are slash commands for complex workflows:\n\n| Skill | What It Does |\n|-------|-------------|\n| `/blog-post` | Multi-agent blog post production pipeline with research and MDX generation |\n| `/cmux` | Terminal CLI reference for cmux multiplexer and session management |\n| `/content-validation` | Validate content integrity beyond HTTP status codes: media, API responses, data contracts |\n| `/deep-research` | Multi-source deep research using Exa, Firecrawl, and WebSearch with citations |\n| `/cross-platform-parsing` | Safe text and CLI output parsing patterns across Windows and Unix |\n| `/game-dev` | Game development team orchestration and project automation |\n| `/gws` | Google Workspace CLI: Drive, Gmail, Calendar, Docs, Sheets, Slides, Tasks, and more |\n| `/homenet-allow-mac` | Add a MAC address to a UniFi SSID's allowlist (preview by default, --apply to commit) |\n| `/homenet-client-profile` | LLM-composed intelligence profile for a single client device joining UniFi state, persona, and Pi-hole DNS evidence (read-only against external systems, writes only to local override table) |\n| `/homenet-deny-mac` | Remove a MAC address from a UniFi SSID's allowlist (preview by default, --apply to commit) |\n| `/homenet-device-profile` | Device-first LAN behavior profile combining UniFi client state and Pi-hole DNS data (read-only) |\n| `/homenet-document` | Generate or refresh comprehensive UniFi network documentation with NotebookLM publication |\n| `/homenet-filter` | Toggle mac_filter_enabled on a UniFi SSID (auto-snapshots before change, refuses empty-allowlist enables) |\n| `/homenet-ppsk-add` | Add a Private Pre-Shared Key (PPSK) entry to a PPSK-enabled SSID (preview by default, --apply to commit) |\n| `/homenet-ppsk-remove` | Remove a PPSK entry from an SSID (preview by default, --apply to commit, refuses to brick SSID) |\n| `/homenet-review` | Reconcile each SSID's MAC allowlist against actually-seen clients (active + historical) |\n| `/homenet-snapshot` | Snapshot all UniFi wlanconf (SSID) state to HomeNetwork/backups for rollback |\n| `/memory-architecture` | Two-tier memory architecture and vector memory configuration for Claude sessions |\n| `/multi-agent-orchestration` | Patterns for structuring multi-agent teams with phase gating and sandbox constraints |\n| `/multi-repo-status` | Git status dashboard across all project repos in parallel |\n| `/notebooklm-content` | Create branded infographics and slide decks from blog posts using Google NotebookLM |\n| `/openclaw-ops` | Configuration gotchas and operational patterns for OpenClaw multi-agent systems |\n| `/skill-catalog` | Full inventory of all agents, skills, commands, and hooks |\n| `/storage-cleanup` | Scan Mac storage, identify cleanup opportunities, and move safe files to external drive |\n| `/sync` | Configuration sync across repos, mirrors local state to git backups |\n| `/ui-ux` | UI/UX design and quality system: aesthetic direction, component architecture, performance, and visual QA with a coordinated agent team |\n| `/wrap-up` | End-of-session wrap-up: update docs, persist to memory systems, commit and push all repos |\n| `/memory-capture-patterns` | Operational patterns for continuous vector memory capture using hooks and save cadence rules |\n| `/vercel-nextjs-debugging` | Debugging patterns for Next.js MDX content and Vercel deployment failures |\n\n**Learned skills** (in `skills/learned/`) are debugging patterns extracted from real sessions. Each documents a non-obvious problem and its solution. 23 unique skills organized into 6 categories:\n\n| Category | Count | Examples |\n|----------|-------|---------|\n| Claude Code | 8 | MCP config location, HEREDOC permission pollution, context compaction |\n| Security | 4 | SSRF prevention, path traversal guards, cookie auth |\n| Next.js | 4 | Client component metadata, MDX sort order, Vercel WAF syntax |\n| Platform | 3 | PowerShell stdin hooks, Git Bash path mangling |\n| Workflow | 2 | Blog post pipeline, parallel agent decomposition |\n| API / Testing | 2 | Anthropic model ID format, Vitest class mock constructor |\n\nSee `skills/learned/INDEX.md` for the full list with descriptions.\n\n### Scripts (27 files)\n\nAutomation scripts in `scripts/` for common operations:\n\n| Script | Purpose |\n|--------|---------|\n| `wrap-up-survey.sh` | Multi-repo wrap-up data collection |\n| `sync-survey.sh` | Config sync status survey |\n| `sync-status.sh` | Quick sync status check across repos |\n| `config-diff.sh` | Compare local config against git repo |\n| `context-health.sh` | Context window health check |\n| `blog-inventory.sh` | Blog post inventory and metadata |\n| `blog-voice-diff.sh` | Compare blog post voice against style guide |\n| `cleanup-session.sh` | Clean up session artifacts |\n| `git-stats.sh` | Git statistics across repos |\n| `validate-mdx.sh` | Validate MDX blog post files |\n| `env.sh` | Shared environment variables for repo paths and tool paths |\n| `memory-maintenance.py` | Memory database maintenance and cleanup |\n| `promote-evolved.sh` | Promote evolved agents, skills, and commands to live config |\n| `bridge-launcher.sh` | Launch the OpenClaw bridge server |\n| `exa-wrapper.sh` | Wrapper to launch the Exa MCP server with secrets loaded from environment |\n| `firecrawl-wrapper.sh` | Wrapper to launch the Firecrawl MCP server with secrets loaded from environment |\n| `memory-toggle.ps1` | Toggle vector memory MCP server on and off (Windows) |\n| `obsidian-wrapper.sh` | Wrapper to launch the Obsidian MCP server with secrets loaded from environment (macOS) |\n| `pihole-wrapper.sh` | Wrapper to launch the Pi-hole MCP server with secrets loaded from environment |\n| `refine-snapshot.sh` | Preserve pre-edit copies of components before /refine applies changes |\n| `unifi-wrapper.sh` | Wrapper to launch the UniFi MCP server with secrets loaded from environment (macOS) |\n| `gmail-metrics-export` | Export Gmail assistant run metrics and session archive to the cryptoflexllc /analytics dashboard |\n\n## MCP Servers (7 configured)\n\nMCP (Model Context Protocol) servers extend Claude Code with capabilities it does not have built in.\n\n| Server | Package | Purpose |\n|--------|---------|---------|\n| **memory** | `@modelcontextprotocol/server-memory` | Knowledge graph for entity relationships |\n| **context7** | `@upstash/context7-mcp` | Live library documentation lookup |\n| **sequential-thinking** | `@modelcontextprotocol/server-sequential-thinking` | Structured multi-step reasoning |\n| **github** | `@modelcontextprotocol/server-github` | GitHub API (issues, PRs, code search) |\n| **project-tools** | Custom (local Node.js) | Repo status, blog tools, session artifacts |\n| **vector-memory** | `mcp-memory-service` (Python) | Hybrid vector + keyword search for long-term memory |\n| **obsidian** | Obsidian MCP plugin | Read/write Obsidian vault files |\n\n### Install Commands\n\n```bash\n# Core servers (recommended for all users)\nclaude mcp add --scope user memory -- npx -y @modelcontextprotocol/server-memory\nclaude mcp add --scope user context7 -- npx -y @upstash/context7-mcp@latest\n\n# Extended reasoning\nclaude mcp add --scope user sequential-thinking -- npx -y @modelcontextprotocol/server-sequential-thinking\n\n# GitHub API access (requires personal access token)\nclaude mcp add-json --scope user github '{\n  \"command\": \"npx\",\n  \"args\": [\"-y\", \"@modelcontextprotocol/server-github\"],\n  \"env\": { \"GITHUB_PERSONAL_ACCESS_TOKEN\": \"YOUR_TOKEN_HERE\" }\n}'\n\n# Project tools (custom, requires local setup)\n# See mcp-servers/README.md for configuration details\ncd mcp-servers/project-tools \u0026\u0026 npm install\n\n# Vector memory (requires Python 3.11+ and Ollama)\n# pip install mcp-memory-service\n# ollama pull nomic-embed-text\n# See mcp-servers/README.md for full setup\n\n# Obsidian (requires Obsidian MCP Tools community plugin)\n# See mcp-servers/README.md for binary path and API key setup\n```\n\nSee [mcp-servers/README.md](./mcp-servers/README.md) for detailed configuration, JSON snippets, and troubleshooting.\n\n## Hooks (12 lifecycle hooks)\n\nHooks in `hooks/` are shell scripts that fire automatically at different points in the Claude Code lifecycle. Configure them in your project's `.claude/settings.local.json` using the template at `hooks/settings.local.json.template`.\n\n| Hook | Event | Purpose |\n|------|-------|---------|\n| `file-guard.sh` | PreToolUse | Block Edit/Write on sensitive files (.env, .pem, credentials) |\n| `kg-update-detect.sh` | PostToolUse | Detect knowledge graph changes and trigger sync reminders |\n| `log-activity.sh` | PostToolUse | Log every tool execution with timestamps to activity log |\n| `memory-checkpoint.sh` | Stop | Structured end-of-session memory checklist across 5 categories |\n| `memory-nudge.sh` | PostToolUse | Remind Claude to save context to vector memory after significant work |\n| `observe-homunculus.sh` | PostToolUse | Capture behavioral observations for the Homunculus learning system |\n| `session-scratchpad.sh` | PostToolUse | Write session state to scratchpad for context recovery |\n| `pre-compact.sh` | PreCompact | Preserve session context before compaction |\n| `prompt-notify.sh` | Stop | Play notification sound when Claude finishes a response |\n| `save-session.sh` | SessionEnd | Archive full conversation transcript on session close |\n| `dispatch.sh` | Multiple | Central dispatcher that routes hook events to other hooks |\n\n**Setup:**\n1. Copy `hooks/` to `~/.claude/hooks/` (or your project's `.claude/hooks/`)\n2. Make scripts executable: `chmod +x hooks/*.sh`\n3. Copy `hooks/settings.local.json.template` to `.claude/settings.local.json`\n4. Replace the `HOOK_COMMAND_*` placeholders with actual paths to your hook scripts\n\n## Homunculus (Continuous Learning System)\n\nThe `homunculus/` directory contains the continuous learning system that extracts behavioral patterns from session transcripts and encodes them as instincts.\n\n```\nhomunculus/\n  identity.json.template    # User identity profile template\n  instincts/\n    personal/                # 50 learned instincts (auto-extracted)\n    inherited/.gitkeep       # Instincts shared from other users\n  evolved/\n    agents/.gitkeep          # Agents evolved from instinct clusters\n    commands/.gitkeep        # Commands evolved from patterns\n    skills/.gitkeep          # Skills evolved from patterns\n```\n\n**How it works:**\n1. The `observe-homunculus.sh` hook captures tool usage observations during sessions\n2. The `skill-extractor` agent processes session transcripts and extracts atomic instincts\n3. Instincts start at 0.4 confidence and increase with repeated evidence\n4. When 3+ instincts cluster in a domain, they can be graduated into learned skills\n\nCurrently contains 50 instincts in `instincts/personal/`, covering patterns from MCP configuration to OpenClaw agent management.\n\n## Templates\n\nThe `templates/` directory contains starter configuration files. Copy them and fill in your values:\n\n| Template | Target Location | Purpose |\n|----------|----------------|---------|\n| `claude.json.template` | `~/.claude.json` | MCP server configuration |\n| `settings.json.template` | `~/.claude/settings.json` | Claude Code settings (model, plugins, permissions) |\n| `env.sh.template` | `~/.claude/scripts/env.sh` | Shared environment variables for scripts |\n| `gitignore.template` | `~/.claude/.gitignore` | Git ignore rules for backing up your config |\n\n## Cross-Platform Support\n\nThis configuration supports both **macOS** and **Windows**:\n\n- **macOS**: Hooks use `.sh` scripts, tools available via Homebrew, no PATH manipulation needed. See `rules/operations/macos-platform.md`.\n- **Windows**: Hooks use `.ps1` scripts with `[Console]::In.ReadToEnd()` for stdin, `cmd /c npx` for MCP servers, OneDrive lock workarounds. See `rules/operations/windows-platform.md`.\n\n## Directory Structure\n\n```\nclaude-code-config/\n  rules/                         # 15 global rule files (4 subdirectories)\n  agents/                        # 34 custom agent definitions\n  skills/                        # 19 invocable skills + 42 learned skills\n  commands/                      # 11 commands\n  scripts/                       # 20 automation scripts\n  hooks/                         # 17 lifecycle hooks (11 macOS/Linux + 6 Windows)\n  mcp-servers/                   # MCP server docs + custom project-tools server\n  templates/                     # Configuration file templates\n  homunculus/                    # Continuous learning system (50 instincts)\n  COMPLETE-GUIDE.md              # Comprehensive beginner walkthrough\n```\n\n## Credits\n\n- **[everything-claude-code](https://github.com/affaan-m/everything-claude-code)** by [Affaan Mustafa](https://github.com/affaan-m): The foundation for the agent orchestration framework and plugin agents. 13 specialized agents, 30+ slash commands, and a plugin system. MIT licensed.\n- **[Claude Code](https://docs.claude.com/en/docs/claude-code)** by [Anthropic](https://www.anthropic.com/): The CLI tool this configuration extends.\n- **[Model Context Protocol](https://github.com/modelcontextprotocol)** community: The MCP server ecosystem.\n- Configuration developed and documented by Chris with Claude.\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fchris2ao%2Fclaude-code-config","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fchris2ao%2Fclaude-code-config","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fchris2ao%2Fclaude-code-config/lists"}