[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"project-72520":3},{"id":4,"name":5,"fullName":6,"owner":7,"repo":5,"description":8,"homepage":9,"htmlUrl":10,"language":11,"languages":10,"totalLinesOfCode":10,"stars":12,"forks":13,"watchers":14,"openIssues":15,"contributorsCount":16,"subscribersCount":16,"size":16,"stars1d":17,"stars7d":18,"stars30d":19,"stars90d":16,"forks30d":16,"starsTrendScore":20,"compositeScore":21,"rankGlobal":10,"rankLanguage":10,"license":22,"archived":23,"fork":23,"defaultBranch":24,"hasWiki":25,"hasPages":23,"topics":26,"createdAt":10,"pushedAt":10,"updatedAt":34,"readmeContent":35,"aiSummary":36,"trendingCount":16,"starSnapshotCount":16,"syncStatus":15,"lastSyncTime":37,"discoverSource":38},72520,"my-claude-code-setup","centminmod\u002Fmy-claude-code-setup","centminmod","Shared starter template configuration and CLAUDE.md memory bank system for Claude Code","https:\u002F\u002Fwww.threads.com\u002F@george_sl_liu",null,"Python",2406,227,29,2,0,8,32,125,24,29.07,"MIT License",false,"master",true,[27,28,29,30,31,32,33],"claude","claude-ai","claude-code","claudecode-config","claudecode-hooks","claudecode-subagents","subagents","2026-06-12 02:03:04","[![GitHub stars](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fcentminmod\u002Fmy-claude-code-setup.svg?style=flat-square)](https:\u002F\u002Fgithub.com\u002Fcentminmod\u002Fmy-claude-code-setup\u002Fstargazers) [![GitHub forks](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fforks\u002Fcentminmod\u002Fmy-claude-code-setup.svg?style=flat-square)](https:\u002F\u002Fgithub.com\u002Fcentminmod\u002Fmy-claude-code-setup\u002Fnetwork) [![GitHub issues](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fissues\u002Fcentminmod\u002Fmy-claude-code-setup.svg?style=flat-square)](https:\u002F\u002Fgithub.com\u002Fcentminmod\u002Fmy-claude-code-setup\u002Fissues)\n\n> **May 4, 2026 — CLAUDE.md template updated.** The `CLAUDE.md` template has been modernized with 3 new variants (`CLAUDE-template-1.md`, `CLAUDE-template-2.md`, `CLAUDE-template-3.md`) following official Anthropic best practices — including dual-memory architecture, progressive disclosure, and standalone behavioral rules. Existing users: use [`CLAUDE-migrate-to-new-template.md`](CLAUDE-migrate-to-new-template.md) as an AI prompt to migrate your current CLAUDE.md to any of the new templates while preserving all your project-specific content. See the [CLAUDE.md Templates](#claudemd-templates) section for details.\n\n* My AI journey Substack to follow me in more indepth AI adventures - https:\u002F\u002Fai.georgeliu.com\n* Threads - https:\u002F\u002Fwww.threads.com\u002F@george_sl_liu\n* BlueSky - https:\u002F\u002Fbsky.app\u002Fprofile\u002Fgeorgesl.bsky.social\n\n# My Claude Code Project's Starter Settings\n\n## Alternate Read Me Guides\n\nThe beauty of using AI is that I can get AI to generate alternate Read Me guides using different formats and styles to suit different readers and learners. So I asked Claude Code to generate alternate Read Me guides in different styles - pick one that suits you best or read all of them to get a better understanding of how to use this project.\n\n* [README.md](README.md) - Original written version I did myself below\n* [README-v2.md](README-v2.md) - Progressive Disclosure Guide (Beginner → Intermediate → Advanced sections)\n* [README-v3.md](README-v3.md) - Task-Based Guide (\"I want to...\" navigation for goal-oriented users)\n* [README-v4.md](README-v4.md) - Technical Reference Manual (Chapter-based, dense reference format) \n\n## Overview\n\nMy Claude Code project's starter settings and Claude Code hooks and slash commands are provided in this repository for users to try out. The [CLAUDE.md](https:\u002F\u002Fgithub.com\u002Fcentminmod\u002Fmy-claude-code-setup\u002Fblob\u002Fmaster\u002FCLAUDE.md) is setup as set of memory bank files to better retain context over many chat sessions. The repo now includes [3 CLAUDE.md templates](#claudemd-templates) following official Anthropic best practices with progressive disclosure, dual-memory architecture, and a [migration guide](CLAUDE-migrate-to-new-template.md) for existing users. Be sure to read the official Claude Code docs first at \u003Chttps:\u002F\u002Fdocs.anthropic.com\u002Fen\u002Fdocs\u002Fclaude-code\u002Foverview> and sign up for a [paid Claude AI account](https:\u002F\u002Fclaude.ai\u002F) to use Claude Code. You can pay for Claude Pro $20\u002Fmonth, Claude Max $100\u002Fmonth or Claude Max $200\u002Fmonth. The paid Claude tier plans will include varying quotas for usage and rate limits outlined [here](https:\u002F\u002Fsupport.anthropic.com\u002Fen\u002Farticles\u002F9797557-usage-limit-best-practices). You can also use Claude Code with [Z.AI](#using-zai-with-claude-code) to get higher token usage quotas and get access to Z.AI GLM-4.7 LLM models within Claude Code. Use [Z.AI invite code for additional 10% discount](https:\u002F\u002Fz.ai\u002Fsubscribe?ic=WWB8IFLROM) which can stack with current 50-60% yearly discounts.\n\n1. Copy the files in this Github repo to your project directory (where you intended codebase will be).\n2. Modify the template files and CLAUDE.md`to your liking. `.claude\u002Fsettings.json` needs to install Terminal-Notifier for macOS https:\u002F\u002Fgithub.com\u002Fcentminmod\u002Fterminal-notifier-setup. If you're not using macOS, you can remove `.claude\u002Fsettings.json`.\n3. After launching Claude Code for the first time within your project directory, run `\u002Finit` so that Claude Code analyses your code base and then populates your memory bank system files as per CLAUDE.md` instructions.\n4. Optional step highly recommended: Install Visual Studio Code ([beginners YouTube video guide](https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=rPITZvwyoMc) and [here](https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=P-5bWpUbO60)) and [Claude Code VSC Extension](https:\u002F\u002Fmarketplace.visualstudio.com\u002Fitems?itemName=anthropic.claude-code).\n5. Optional step highly recommended: Sign up for [Github.com](https:\u002F\u002Fgithub.com\u002F) account and install Git for Visual Studio Code. Checkout YouTube guides [here](https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=twsYxYaQikI) and [here](https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=z5jZ9lrSpqk).\n6. CLAUDE.md updated to instruct models to use faster tools so for macOS: `brew install ripgrep fd jq`\n7. Optional step to setup Claude Code, Codex GPT-5, Gemini CLI, OpenCode, Vercel CLI, Cloudflare Wrangler, Amazon AWS CLI, all in a single isolated [Visual Studio Code dev container running Debian 12](https:\u002F\u002Fclaude-devcontainers.centminmod.com\u002F). Allowing you to run YOLO modes for Claude Code with `dangerously_skip_permissions` enabled and Codex CLI with `sandbox_mode = danger-full-access` etc.\n8. Claude Code via Claude Desktop apps use Git Worktrees. You may need to create a `.worktreeinclude` file as outlined [here](https:\u002F\u002Fcode.claude.com\u002Fdocs\u002Fen\u002Fdesktop#claude-code-on-desktop-preview).\n9. If you use Cloudflare and ClerkOS platforms in your apps, you can keep either `CLAUDE-cloudflare.md` or `CLAUDE-cloudflare-mini.md` supplementary reference docs and update `CLAUDE.md` referencing either file to help AI understand Cloudflare and ClerkOS platforum documentation and products. Edit documentation templates as needed i.e. if you do not use ClerkOS platform, you can remove those sections.\n10. If you use Convex database in your apps, you can use `CLAUDE-convex.md` supplementary reference docs for building Next.js and React apps with Convex backend deployed on Cloudflare Pages.\n11. Useful read [Advent of Claude: 31 Days of Claude Code by Ado Kukic from Anthropic](https:\u002F\u002Fadocomplete.com\u002Fadvent-of-claude-2025\u002F).\n12. Useful [Claude Code with Claude Opus 4.5 YouTube video by Alex Finn](https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=UVJXh57MgI0) and [Claude Code YouTube video by Matt Maher](https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=Dekx_OzRwiI).\n13. Configure Claude Code with [Z.AI](#using-zai-with-claude-code) to get higher token usage quotas and get access to Z.AI GLM-4.7 LLM models within Claude Code.\n14. For folks intending to use Claude LLM models with OpenClaw, read this first for security, deployment and cost optimizations https:\u002F\u002Fgithub.com\u002Fcentminmod\u002Fexplain-openclaw.\n\n## CLAUDE.md Memory Bank system\n\n[CLAUDE.md](https:\u002F\u002Fgithub.com\u002Fcentminmod\u002Fmy-claude-code-setup\u002Fblob\u002Fmaster\u002FCLAUDE.md) uses a memory bank system of files for Claude Code to better retain context over many chat sessions. Example of Claude Code thinking output when I ask it to `update memory bank` after a successful task completion and git commit:\n\n```\n> update memory bank\n\n∴ Thinking…\n\n  The user wants me to update the memory bank files. I should update the relevant CLAUDE-*.md files to reflect the v0.1.6 changes and current state. Let me check\n   which memory bank files exist and update them.\n\n∴ Thinking…\n\n  I need to continue updating the memory bank files. I'll update:\n  1. CLAUDE-patterns.md - Add export path handling pattern (Pattern 22)\n  2. CLAUDE-decisions.md - Add ADR-017 for Phoneme API migration and ADR-018 for export path handling\n  3. CLAUDE-troubleshooting.md - Add new issues and solutions for v0.1.6\n\n∴ Thinking…\n\n  The memory bank files have been modified but not committed. The v0.1.6 release was already committed and tagged, but the memory bank documentation updates are\n  still uncommitted. Let me provide a summary of the completed work and offer to commit these documentation changes.\n\n⏺ Summary: v0.1.6 Release Complete ✅\n\n  Memory Bank Updates Complete:\n  - CLAUDE-activeContext.md - v0.1.6 session context\n  - CLAUDE-patterns.md - Added patterns 22-24 (Export Path, Format Detection, API Migration)\n  - CLAUDE-decisions.md - Added ADR-017 & ADR-018\n  - CLAUDE-troubleshooting.md - Added 4 new issue\u002Ffix entries\n\n  v0.1.6 Fixes Applied:\n  1. Export to absolute paths (e.g., --export \u002Ftmp\u002Ftest.wav)\n  2. Format detection from filename extension (prevents .mp3.wav)\n  3. Phoneme API migrated to \u002Fdev\u002Fphonemize\n  4. Test scripts updated for --export \"\" syntax\n\n  Test Results: All 29 tests passing\n\n  The memory bank documentation updates are ready to commit. Would you like me to commit them\n```\n\n## CLAUDE.md Templates\n\nThis repo provides 3 modernized CLAUDE.md templates following [official Anthropic best practices](https:\u002F\u002Fcode.claude.com\u002Fdocs\u002Fen\u002Fmemory):\n\n| Template | Lines | Philosophy | Best For |\n|----------|-------|------------|----------|\n| [`CLAUDE-template-1.md`](CLAUDE-template-1.md) | ~101 | Compact self-contained + memory resilience | Quick starts, small projects |\n| [`CLAUDE-template-2.md`](CLAUDE-template-2.md) | ~153 | Memory bank headline + dual memory | Existing memory bank users |\n| [`CLAUDE-template-3.md`](CLAUDE-template-3.md) | ~105 | Progressive disclosure native | Teams, max context efficiency |\n\nThe current `CLAUDE.md` uses the Template 3 format. All templates include:\n- Dual-memory architecture (git-shared CLAUDE-*.md + machine-local auto memory for resilience)\n- Progressive disclosure (memory bank files read on-demand, not auto-loaded)\n- Standalone behavioral rules (no dependency on `~\u002F.claude\u002FCLAUDE.md`)\n\n**Migrating an existing CLAUDE.md?** See [`CLAUDE-migrate-to-new-template.md`](CLAUDE-migrate-to-new-template.md) — paste it as a prompt and Claude Code will migrate your content to your chosen template format.\n\n**Tip:** HTML comments (`\u003C!-- -->`) in CLAUDE.md are stripped from context at runtime and cost zero tokens. The templates use them for guidance notes — you can add your own without impacting Claude's context window.\n\n**Companion rules file:** Template 3 requires [`.claude\u002Frules\u002Fcore-rules.md`](.claude\u002Frules\u002Fcore-rules.md) for behavioral rules. Templates 1 and 2 have rules inline but can optionally use it.\n\n## MCP Servers\n\nI also install the following MCP servers ([install commands](#claude-code-mcp-servers)):\n\n* [Gemini CLI MCP](https:\u002F\u002Fgithub.com\u002Fcentminmod\u002Fgemini-cli-mcp-server)\n* [Cloudflare Documentation MCP](https:\u002F\u002Fgithub.com\u002Fcloudflare\u002Fmcp-server-cloudflare\u002Ftree\u002Fmain\u002Fapps\u002Fdocs-vectorize)\n* [Context 7 MCP](https:\u002F\u002Fgithub.com\u002Fupstash\u002Fcontext7)\n* [Chrome Devtools MCP](https:\u002F\u002Fgithub.com\u002FChromeDevTools\u002Fchrome-devtools-mcp)\n* [Notion MCP](https:\u002F\u002Fgithub.com\u002Fmakenotion\u002Fnotion-mcp-server)\n* [Claude Code Usage Metrics MCP](https:\u002F\u002Fgithub.com\u002Fcentminmod\u002Fclaude-code-opentelemetry-setup)\n\n## Claude Code Plugin Marketplace\n\nBrowser and install Claude Code plugins from official marketplace https:\u002F\u002Fcode.claude.com\u002Fdocs\u002Fen\u002Fdiscover-plugins.\n\n```bash\n\u002Fplugin install plugin-name@claude-plugins-official\n```\n\nTo update official Claude Code plugin marketplace:\n\n```bash\n\u002Fplugin marketplace update claude-plugins-official\n```\n\nInstall [code simplifier plugin](https:\u002F\u002Fgithub.com\u002Fanthropics\u002Fclaude-plugins-official\u002Ftree\u002Fmain\u002Fplugins\u002Fcode-simplifier).\n\n> Agent that simplifies and refines code for clarity, consistency, and maintainability while preserving functionality. Focuses on recently modified code.\n\n```bash\n\u002Fplugin install code-simplifier\n```\n\nInstall [frontend design plugin](https:\u002F\u002Fgithub.com\u002Fanthropics\u002Fclaude-code\u002Ftree\u002Fmain\u002Fplugins\u002Ffrontend-design).\n\n> Create distinctive, production-grade frontend interfaces with high design quality. Generates creative, polished code that avoids generic AI aesthetics.\n\n```bash\n\u002Fplugin install frontend-design@claude-code-plugins\n```\n\nInstall [feature dev plugin](https:\u002F\u002Fgithub.com\u002Fanthropics\u002Fclaude-code\u002Ftree\u002Fmain\u002Fplugins\u002Ffeature-dev).\n\n> The Feature Development Plugin provides a systematic 7-phase approach to building new features. Instead of jumping straight into code, it guides you through understanding the codebase, asking clarifying questions, designing architecture, and ensuring quality—resulting in better-designed features that integrate seamlessly with your existing code.\n\n```bash\n\u002Fplugin install feature-dev@claude-code-plugins\n```\n\nInstall [Ralph Wiggum plugin](https:\u002F\u002Fgithub.com\u002Fanthropics\u002Fclaude-code\u002Ftree\u002Fmain\u002Fplugins\u002Fralph-wiggum). Details https:\u002F\u002Fpaddo.dev\u002Fblog\u002Fralph-wiggum-autonomous-loops\u002F. Update: currently seems broken on some systems.\n\n- https:\u002F\u002Fgithub.com\u002Fanthropics\u002Fclaude-code\u002Fissues\u002F16398\n- https:\u002F\u002Fgithub.com\u002Fanthropics\u002Fclaude-code\u002Fissues\u002F16389\n\n> Interactive self-referential AI loops for iterative development. Claude works on the same task repeatedly, seeing its previous work, until completion.\n\n```bash\n\u002Fplugin install ralph-wiggum@claude-code-plugins\n```\n\nHow to use Ralph Wiggum plugin YouTube by Greg Isenberg and Ryan Carson https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=RpvQH0r0ecM and GitHub repo https:\u002F\u002Fgithub.com\u002Fsnarktank\u002Fralph.\n\n## Claude Code 3rd Party Marketplaces\n\nClaude Code Safety Net plugin https:\u002F\u002Fgithub.com\u002Fkenryu42\u002Fclaude-code-safety-net to prevent destructive commands being run by Claude Code i.e. https:\u002F\u002Fwww.reddit.com\u002Fr\u002FClaudeAI\u002Fcomments\u002F1pgxckk\u002Fclaude_cli_deleted_my_entire_home_directory_wiped\u002F\n\n> A Claude Code plugin that acts as a safety net, catching destructive git and filesystem commands before they execute.\n\n```bash\n\u002Fplugin marketplace add kenryu42\u002Fcc-marketplace\n\u002Fplugin install safety-net@cc-marketplace\n```\n\n[Z.AI usage query plugin](https:\u002F\u002Fdocs.z.ai\u002Fdevpack\u002Fextension\u002Fusage-query-plugin) for querying [Z.AI](#using-zai-with-claude-code) usage statistics.\n\n> Query your current quota and usage statistics for the GLM Coding Plan directly within Claude Code.\n\n```bash\n\u002Fplugin marketplace add zai\u002Fzai-coding-plugins\n\u002Fplugin install glm-plan-usage@zai-coding-plugins\n```\n\n[Cloudflare Skills marketplace](https:\u002F\u002Fgithub.com\u002Fcloudflare\u002Fskills) for building applications on Cloudflare's platform, Workers, and the Agents SDK.\n\n> Collection of Agent Skills providing accurate, up-to-date guidance for Cloudflare development tasks including Workers, Pages, AI services, and infrastructure.\n\n```bash\n\u002Fplugin marketplace add cloudflare\u002Fskills\n```\n\n**User commands**: `\u002Fcloudflare:build-agent`, `\u002Fcloudflare:build-mcp`\n\n## Claude Code Statuslines\n\n`~\u002F.claude\u002Fstatuslines\u002Fstatusline.sh` configured in `~\u002F.claude\u002Fsettings.json`.\n\nfor `~\u002F.claude\u002Fsettings.json`\n\n```bash\n  \"statusLine\": {\n    \"type\": \"command\",\n    \"command\": \"~\u002F.claude\u002Fstatuslines\u002Fstatusline.sh\",\n    \"padding\": 0\n  },\n```\n\nfor `~\u002F.claude\u002Fstatuslines\u002Fstatusline.sh`\n\n```bash\n#!\u002Fbin\u002Fbash\n# Read JSON input from stdin\ninput=$(cat)\n\n# Extract model and workspace values\nMODEL_DISPLAY=$(echo \"$input\" | jq -r '.model.display_name')\nCURRENT_DIR=$(echo \"$input\" | jq -r '.workspace.current_dir')\n\n# Extract context window metrics\nINPUT_TOKENS=$(echo \"$input\" | jq -r '.context_window.total_input_tokens')\nOUTPUT_TOKENS=$(echo \"$input\" | jq -r '.context_window.total_output_tokens')\nCONTEXT_SIZE=$(echo \"$input\" | jq -r '.context_window.context_window_size')\n\n# Extract cost metrics\nCOST_USD=$(echo \"$input\" | jq -r '.cost.total_cost_usd')\nLINES_ADDED=$(echo \"$input\" | jq -r '.cost.total_lines_added')\nLINES_REMOVED=$(echo \"$input\" | jq -r '.cost.total_lines_removed')\n\n# Extract percentage metrics\nUSED_PERCENTAGE=$(echo \"$input\" | jq -r '.context_window.used_percentage')\nREMAINING_PERCENTAGE=$(echo \"$input\" | jq -r '.context_window.remaining_percentage')\n\n# Format tokens as Xk\nformat_tokens() {\n    local num=\"$1\"\n    if [ \"$num\" -ge 1000 ]; then\n        echo \"$((num \u002F 1000))k\"\n    else\n        echo \"$num\"\n    fi\n}\n\n# Generate progress bar for context usage\ngenerate_progress_bar() {\n    local percentage=$1\n    local bar_width=20\n    local filled=$(awk \"BEGIN {printf \\\"%.0f\\\", ($percentage \u002F 100) * $bar_width}\")\n    local empty=$((bar_width - filled))\n    local bar=\"\"\n    for ((i=0; i\u003Cfilled; i++)); do bar+=\"█\"; done\n    for ((i=0; i\u003Cempty; i++)); do bar+=\"░\"; done\n    echo \"$bar\"\n}\n\n# Calculate total\nTOTAL_TOKENS=$((INPUT_TOKENS + OUTPUT_TOKENS))\n\n# Generate progress bar\nPROGRESS_BAR=$(generate_progress_bar \"$USED_PERCENTAGE\")\n\n# Show git branch if in a git repo\nGIT_BRANCH=\"\"\nif git -C \"$CURRENT_DIR\" rev-parse --git-dir > \u002Fdev\u002Fnull 2>&1; then\n    BRANCH=$(git -C \"$CURRENT_DIR\" branch --show-current 2>\u002Fdev\u002Fnull)\n    if [ -n \"$BRANCH\" ]; then\n        # Worktree detection\n        GIT_DIR=$(git -C \"$CURRENT_DIR\" rev-parse --git-dir 2>\u002Fdev\u002Fnull)\n        WORKTREE=\"\"\n        if [[ \"$GIT_DIR\" == *\".git\u002Fworktrees\u002F\"* ]] || [[ -f \"$GIT_DIR\u002Fgitdir\" ]]; then\n            WORKTREE=\" 🌳\"\n        fi\n        # Ahead\u002Fbehind detection\n        AHEAD_BEHIND=\"\"\n        UPSTREAM=$(git -C \"$CURRENT_DIR\" rev-parse --abbrev-ref '@{u}' 2>\u002Fdev\u002Fnull)\n        if [ -n \"$UPSTREAM\" ]; then\n            AHEAD=$(git -C \"$CURRENT_DIR\" rev-list --count '@{u}..HEAD' 2>\u002Fdev\u002Fnull || echo 0)\n            BEHIND=$(git -C \"$CURRENT_DIR\" rev-list --count 'HEAD..@{u}' 2>\u002Fdev\u002Fnull || echo 0)\n            if [ \"$AHEAD\" -gt 0 ] && [ \"$BEHIND\" -gt 0 ]; then\n                AHEAD_BEHIND=\" ↕${AHEAD}\u002F${BEHIND}\"\n            elif [ \"$AHEAD\" -gt 0 ]; then\n                AHEAD_BEHIND=\" ↑${AHEAD}\"\n            elif [ \"$BEHIND\" -gt 0 ]; then\n                AHEAD_BEHIND=\" ↓${BEHIND}\"\n            fi\n        fi\n        GIT_BRANCH=\" | 🌿 $BRANCH${WORKTREE}${AHEAD_BEHIND}\"\n    fi\nfi\n\necho \"[$MODEL_DISPLAY] 📁 ${CURRENT_DIR##*\u002F}${GIT_BRANCH}\nContext: [$PROGRESS_BAR] ${USED_PERCENTAGE}%\nCost: \\$${COST_USD} | +${LINES_ADDED} -${LINES_REMOVED} lines\"\n```\n\n## Git Worktrees for AI Coding Sessions\n\nGit worktrees allow you to run parallel Claude Code and Codex CLI sessions with complete code isolation. Each worktree has its own isolated working directory while sharing the same Git history and remote connections. This prevents AI instances from interfering with each other when working on multiple tasks simultaneously.\n\n**Benefits:**\n- Run multiple AI coding sessions in parallel\n- Each worktree has independent file state\n- Changes in one worktree won't affect others\n- Ideal for experimental features or YOLO mode usage\n\n**Official Documentation:** [Run parallel Claude Code sessions with git worktrees](https:\u002F\u002Fcode.claude.com\u002Fdocs\u002Fen\u002Fcommon-workflows#run-parallel-claude-code-sessions-with-git-worktrees)\n\n### macOS \u002F Linux (Bash\u002FZsh)\n\nAdd these functions to `~\u002F.bashrc`, `~\u002F.zshrc`, or `~\u002F.bash_aliases`:\n\n```bash\n# Codex CLI worktree launcher\ncx() {\n    local branch_name\n    if [ -z \"$1\" ]; then\n        branch_name=\"worktree-$(date +%Y%m%d-%H%M%S)\"\n    else\n        branch_name=\"$1\"\n    fi\n    git worktree add \"..\u002F$branch_name\" -b \"$branch_name\" && \\\n    cd \"..\u002F$branch_name\" || return 1\n    codex -m gpt-5-codex --config model_reasoning_effort='xhigh'\n}\n\n# Claude Code worktree launcher\nclx() {\n    local branch_name\n    if [ -z \"$1\" ]; then\n        branch_name=\"worktree-$(date +%Y%m%d-%H%M%S)\"\n    else\n        branch_name=\"$1\"\n    fi\n    git worktree add \"..\u002F$branch_name\" -b \"$branch_name\" && \\\n    cd \"..\u002F$branch_name\" || return 1\n    claude --model opusplan --permission-mode plan\n}\n```\n\nAfter adding, reload your shell: `source ~\u002F.bashrc` or `source ~\u002F.zshrc`\n\n### Windows (PowerShell)\n\nAdd these functions to your PowerShell profile. Open it with `notepad $PROFILE`:\n\n```powershell\n# Codex CLI worktree launcher\nfunction cx {\n    param([string]$BranchName)\n    if (-not $BranchName) {\n        $BranchName = \"worktree-$(Get-Date -Format 'yyyyMMdd-HHmmss')\"\n    }\n    git worktree add \"..\u002F$BranchName\" -b $BranchName\n    if ($LASTEXITCODE -eq 0) {\n        Set-Location \"..\u002F$BranchName\"\n        codex -m gpt-5-codex --config model_reasoning_effort='xhigh'\n    }\n}\n\n# Claude Code worktree launcher\nfunction clx {\n    param([string]$BranchName)\n    if (-not $BranchName) {\n        $BranchName = \"worktree-$(Get-Date -Format 'yyyyMMdd-HHmmss')\"\n    }\n    git worktree add \"..\u002F$BranchName\" -b $BranchName\n    if ($LASTEXITCODE -eq 0) {\n        Set-Location \"..\u002F$BranchName\"\n        claude --model opusplan --permission-mode plan\n    }\n}\n```\n\nAfter adding, reload PowerShell or run: `. $PROFILE`\n\n### Windows (CMD Batch Files)\n\nCreate these batch files in a directory in your PATH (e.g., `C:\\Users\\YourName\\bin\\`):\n\n**cx.bat** - Codex CLI launcher:\n\n```batch\n@echo off\nsetlocal enabledelayedexpansion\nif \"%~1\"==\"\" (\n    for \u002Ff \"tokens=2 delims==\" %%I in ('wmic os get localdatetime \u002Fvalue') do set datetime=%%I\n    set branch_name=worktree-!datetime:~0,8!-!datetime:~8,6!\n) else (\n    set branch_name=%~1\n)\ngit worktree add \"..\u002F%branch_name%\" -b \"%branch_name%\"\nif %errorlevel% equ 0 (\n    cd \"..\u002F%branch_name%\"\n    codex -m gpt-5-codex --config model_reasoning_effort='xhigh'\n)\nendlocal\n```\n\n**clx.bat** - Claude Code launcher:\n\n```batch\n@echo off\nsetlocal enabledelayedexpansion\nif \"%~1\"==\"\" (\n    for \u002Ff \"tokens=2 delims==\" %%I in ('wmic os get localdatetime \u002Fvalue') do set datetime=%%I\n    set branch_name=worktree-!datetime:~0,8!-!datetime:~8,6!\n) else (\n    set branch_name=%~1\n)\ngit worktree add \"..\u002F%branch_name%\" -b \"%branch_name%\"\nif %errorlevel% equ 0 (\n    cd \"..\u002F%branch_name%\"\n    claude --model opusplan --permission-mode plan\n)\nendlocal\n```\n\n### Usage\n\n```bash\n# Create worktree with custom name\nclx feature-auth\ncx bugfix-123\n\n# Create worktree with auto-generated timestamp name\nclx\ncx\n```\n\n### Worktree Management\n\n```bash\n# List all worktrees\ngit worktree list\n\n# Remove a worktree when done\ngit worktree remove ..\u002Fworktree-name\n\n# Clean up stale worktree references\ngit worktree prune\n```\n\n### Environment Setup\n\nEach new worktree needs its own development environment:\n- **JavaScript\u002FNode.js**: Run `npm install` or `yarn`\n- **Python**: Create virtual environment or run `pip install -r requirements.txt`\n- **Other languages**: Follow your project's standard setup process\n\n### The `.worktreeinclude` File\n\nWhen Claude Code creates a worktree, files ignored via `.gitignore` aren't automatically available. The `.worktreeinclude` file specifies which ignored files should be copied to new worktrees.\n\n**How It Works:**\n- Uses `.gitignore`-style patterns\n- Only files matched by **both** `.worktreeinclude` **AND** `.gitignore` are copied\n- This prevents accidentally duplicating tracked files\n\nCreate a `.worktreeinclude` file in your repository root:\n\n```text\n# Environment files\n.env\n.env.local\n.env.*\n\n# Claude Code local settings\n**\u002F.claude\u002Fsettings.local.json\n```\n\n**Common Use Cases:**\n- `.env` files with API keys and secrets\n- `.env.local` for local development overrides\n- `.claude\u002Fsettings.local.json` for personal Claude Code settings\n\n### Claude Desktop Worktree Location\n\nWhen using Claude Code via the Claude Desktop app:\n- Default worktree location: `~\u002F.claude-worktrees`\n- Configurable through Claude Desktop app settings\n- Repository must be Git initialized for worktree sessions to work\n\n**Official Documentation:** [Claude Code on Desktop](https:\u002F\u002Fcode.claude.com\u002Fdocs\u002Fen\u002Fdesktop#claude-code-on-desktop-preview)\n\n### Local Ignores Without `.gitignore`\n\nTo ignore files locally without modifying the shared `.gitignore`, use `.git\u002Finfo\u002Fexclude`:\n\n```bash\n# Edit the local exclude file\nnano .git\u002Finfo\u002Fexclude\n# or\ncode .git\u002Finfo\u002Fexclude\n```\n\nAdd patterns using the same syntax as `.gitignore`:\n\n```text\n# Local IDE settings\n.idea\u002F\n*.swp\n\n# Personal scripts\nmy-local-scripts\u002F\n\n# Local test files\ntest-local.sh\n```\n\n**Key Differences:**\n\n| File | Scope | Committed to Git |\n|------|-------|------------------|\n| `.gitignore` | Shared with team | Yes |\n| `.git\u002Finfo\u002Fexclude` | Local only | No |\n| `~\u002F.config\u002Fgit\u002Fignore` | Global (all repos) | No |\n\n**When to Use `.git\u002Finfo\u002Fexclude`:**\n- Personal IDE or editor files\n- Local testing scripts\n- Machine-specific configurations\n- Files you don't want to clutter the shared `.gitignore`\n\n**Note:** Files in `.git\u002Finfo\u002Fexclude` work with `.worktreeinclude` the same way as `.gitignore` - patterns must appear in both files for copying to worktrees.\n\n## Using Z.AI with Claude Code\n\n* Get extra 10% discount with invite code [(https:\u002F\u002Fz.ai\u002Fsubscribe?ic=WWB8IFLROM)](https:\u002F\u002Fz.ai\u002Fsubscribe?ic=WWB8IFLROM)\n\nZ.AI's GLM Coding Plan is a cost-effective subscription service starting at just $3\u002Fmonth that provides access to GLM-4.7, a high-performance language model optimized for coding tasks. With over 55 tokens per second for real-time interaction, it offers state-of-the-art performance in reasoning, coding, and agent capabilities. The service includes multimodal features like Vision Understanding, Web Search, and Web Reader MCP servers. Below [shell function launchers](#shell-function-launchers) are easiest way to use Z.AI with Claude Code without messing up your existing Claude Code setup.\n\n**Usage Tiers:**\n- **Lite Plan** (~$3\u002Fmonth): ~120 prompts per 5 hours (roughly 3× Claude Pro quota)\n- **Pro Plan**: ~600 prompts per 5 hours (3× Claude Max 5x quota)\n- **Max Plan**: ~2,400 prompts per 5 hours (3× Claude Max 20x quota)\n\nEach prompt typically allows 15–20 model calls, yielding tens of billions of tokens monthly at approximately 1% of standard API pricing.\n\n### Privacy & Data Handling\n\n**Data Location:** All Z.AI services are based in Singapore.\n\n**Privacy Guarantee:** Z.AI does not store any of the content you provide or generate while using their services. This includes any text prompts, images, or other data you input.\n\nSee the [Privacy Policy](https:\u002F\u002Fdocs.z.ai\u002Flegal-agreement\u002Fprivacy-policy) for further details.\n\n### Special Discount: Get 10% OFF!\n\n> 🚀 You've been invited to join the GLM Coding Plan! Enjoy full support for Claude Code, Cline, and 10+ top coding tools — starting at just $3\u002Fmonth. Subscribe now and grab the limited-time deal!\n>\n> **Invite Code:** `WWB8IFLROM` (10% additional discount)\n>\n> **Subscribe:** [https:\u002F\u002Fz.ai\u002Fsubscribe?ic=WWB8IFLROM](https:\u002F\u002Fz.ai\u002Fsubscribe?ic=WWB8IFLROM)\n\n### Prerequisites\n\n- Node.js 18 or newer\n- Z.AI API key (obtain from [Z.AI dashboard](https:\u002F\u002Fz.ai))\n- Official documentation: [https:\u002F\u002Fdocs.z.ai\u002Fdevpack\u002Ftool\u002Fclaude](https:\u002F\u002Fdocs.z.ai\u002Fdevpack\u002Ftool\u002Fclaude)\n\n### Setup Instructions\n\n#### Option 1: Automated Setup (Recommended for First-Time Users)\n\nJust run the following command in your terminal. Attention only macOS Linux environment is supported, this method does not support Windows\n\n```bash\ncurl -O \"https:\u002F\u002Fcdn.bigmodel.cn\u002Finstall\u002Fclaude_code_zai_env.sh\" && bash .\u002Fclaude_code_zai_env.sh\n```\n\n#### Option 2: Manual Configuration\n\nEdit your `~\u002F.claude\u002Fsettings.json` and add the environment variables:\n\n```json\n{\n  \"env\": {\n    \"ANTHROPIC_AUTH_TOKEN\": \"your-zai-api-key\",\n    \"ANTHROPIC_BASE_URL\": \"https:\u002F\u002Fapi.z.ai\u002Fapi\u002Fanthropic\",\n    \"API_TIMEOUT_MS\": \"3000000\"\n  }\n}\n```\n\nReplace `your-zai-api-key` with your actual Z.AI API key from the dashboard.\n\n### Shell Function Launchers\n\n#### macOS \u002F Linux (Bash\u002FZsh)\n\nAdd this function to `~\u002F.bashrc`, `~\u002F.zshrc`, or `~\u002F.bash_aliases`:\n\n```bash\n# Z.AI + Claude Code launcher\nzai() {\n    export ANTHROPIC_AUTH_TOKEN=\"your-zai-api-key\"\n    export ANTHROPIC_BASE_URL=\"https:\u002F\u002Fapi.z.ai\u002Fapi\u002Fanthropic\"\n    export API_TIMEOUT_MS=\"3000000\"\n    claude \"$@\"\n}\n```\n\nAfter adding, reload your shell: `source ~\u002F.bashrc` or `source ~\u002F.zshrc`\n\n#### Windows (PowerShell)\n\nAdd this function to your PowerShell profile. Open it with `notepad $PROFILE`:\n\n```powershell\n# Z.AI + Claude Code launcher\nfunction zai {\n    $env:ANTHROPIC_AUTH_TOKEN = \"your-zai-api-key\"\n    $env:ANTHROPIC_BASE_URL = \"https:\u002F\u002Fapi.z.ai\u002Fapi\u002Fanthropic\"\n    $env:API_TIMEOUT_MS = \"3000000\"\n    claude $args\n}\n```\n\nAfter adding, reload PowerShell or run: `. $PROFILE`\n\n#### Windows (CMD Batch Files)\n\nCreate a batch file named `zai.bat` in a directory in your PATH (e.g., `C:\\Users\\YourName\\bin\\`):\n\n```batch\n@echo off\nset ANTHROPIC_AUTH_TOKEN=your-zai-api-key\nset ANTHROPIC_BASE_URL=https:\u002F\u002Fapi.z.ai\u002Fapi\u002Fanthropic\nset API_TIMEOUT_MS=3000000\nclaude %*\n```\n\n### Model Mapping\n\nThe GLM Coding Plan uses intelligent model mapping where Claude model names are mapped to GLM models through environment variables:\n\n- `ANTHROPIC_DEFAULT_OPUS_MODEL`: GLM-4.7\n- `ANTHROPIC_DEFAULT_SONNET_MODEL`: GLM-4.7\n- `ANTHROPIC_DEFAULT_HAIKU_MODEL`: GLM-4.5-Air\n\n**Note:** While manual adjustment of these mappings is possible, it's not recommended as it may prevent automatic updates to newer model versions.\n\n#### Customizing Model Mappings\n\nIf you need to switch to different models (e.g., GLM-4.5 or other models), you can customize the mappings in two ways:\n\n**Option 1: In `~\u002F.claude\u002Fsettings.json`**\n\n```json\n{\n  \"env\": {\n    \"ANTHROPIC_AUTH_TOKEN\": \"your-zai-api-key\",\n    \"ANTHROPIC_BASE_URL\": \"https:\u002F\u002Fapi.z.ai\u002Fapi\u002Fanthropic\",\n    \"API_TIMEOUT_MS\": \"3000000\",\n    \"ANTHROPIC_DEFAULT_OPUS_MODEL\": \"GLM-4.7\",\n    \"ANTHROPIC_DEFAULT_SONNET_MODEL\": \"GLM-4.5\",\n    \"ANTHROPIC_DEFAULT_HAIKU_MODEL\": \"GLM-4.5-Air\"\n  }\n}\n```\n\n**Option 2: In Shell Function\u002FAlias**\n\nFor macOS\u002FLinux:\n\n```bash\nzai() {\n    export ANTHROPIC_AUTH_TOKEN=\"your-zai-api-key\"\n    export ANTHROPIC_BASE_URL=\"https:\u002F\u002Fapi.z.ai\u002Fapi\u002Fanthropic\"\n    export API_TIMEOUT_MS=\"3000000\"\n    export ANTHROPIC_DEFAULT_OPUS_MODEL=\"GLM-4.7\"\n    export ANTHROPIC_DEFAULT_SONNET_MODEL=\"GLM-4.5\"\n    export ANTHROPIC_DEFAULT_HAIKU_MODEL=\"GLM-4.5-Air\"\n    claude \"$@\"\n}\n```\n\nFor Windows PowerShell:\n\n```powershell\nfunction zai {\n    $env:ANTHROPIC_AUTH_TOKEN = \"your-zai-api-key\"\n    $env:ANTHROPIC_BASE_URL = \"https:\u002F\u002Fapi.z.ai\u002Fapi\u002Fanthropic\"\n    $env:API_TIMEOUT_MS = \"3000000\"\n    $env:ANTHROPIC_DEFAULT_OPUS_MODEL = \"GLM-4.7\"\n    $env:ANTHROPIC_DEFAULT_SONNET_MODEL = \"GLM-4.5\"\n    $env:ANTHROPIC_DEFAULT_HAIKU_MODEL = \"GLM-4.5-Air\"\n    claude $args\n}\n```\n\nFor Windows CMD:\n\n```batch\n@echo off\nset ANTHROPIC_AUTH_TOKEN=your-zai-api-key\nset ANTHROPIC_BASE_URL=https:\u002F\u002Fapi.z.ai\u002Fapi\u002Fanthropic\nset API_TIMEOUT_MS=3000000\nset ANTHROPIC_DEFAULT_OPUS_MODEL=GLM-4.7\nset ANTHROPIC_DEFAULT_SONNET_MODEL=GLM-4.5\nset ANTHROPIC_DEFAULT_HAIKU_MODEL=GLM-4.5-Air\nclaude %*\n```\n\n### Usage\n\n```bash\n# Launch with Z.AI configuration\nzai\n\n# Launch with specific model\nzai --model sonnet\n\n# Launch with permission mode\nzai --model opus --permission-mode plan\n```\n\n### Z.AI + Git Worktree Integration\n\nCombine Z.AI with git worktrees for isolated parallel sessions:\n\n#### macOS \u002F Linux (Bash\u002FZsh)\n\n```bash\n# Z.AI + Claude Code worktree launcher\nzaix() {\n    local branch_name\n    if [ -z \"$1\" ]; then\n        branch_name=\"worktree-$(date +%Y%m%d-%H%M%S)\"\n    else\n        branch_name=\"$1\"\n    fi\n    git worktree add \"..\u002F$branch_name\" -b \"$branch_name\" && \\\n    cd \"..\u002F$branch_name\" || return 1\n\n    export ANTHROPIC_AUTH_TOKEN=\"your-zai-api-key\"\n    export ANTHROPIC_BASE_URL=\"https:\u002F\u002Fapi.z.ai\u002Fapi\u002Fanthropic\"\n    export API_TIMEOUT_MS=\"3000000\"\n    claude --model sonnet --permission-mode plan\n}\n```\n\n#### Windows (PowerShell)\n\n```powershell\n# Z.AI + Claude Code worktree launcher\nfunction zaix {\n    param([string]$BranchName)\n    if (-not $BranchName) {\n        $BranchName = \"worktree-$(Get-Date -Format 'yyyyMMdd-HHmmss')\"\n    }\n    git worktree add \"..\u002F$BranchName\" -b $BranchName\n    if ($LASTEXITCODE -eq 0) {\n        Set-Location \"..\u002F$BranchName\"\n        $env:ANTHROPIC_AUTH_TOKEN = \"your-zai-api-key\"\n        $env:ANTHROPIC_BASE_URL = \"https:\u002F\u002Fapi.z.ai\u002Fapi\u002Fanthropic\"\n        $env:API_TIMEOUT_MS = \"3000000\"\n        claude --model sonnet --permission-mode plan\n    }\n}\n```\n\n#### Windows (CMD Batch Files)\n\nCreate `zaix.bat`:\n\n```batch\n@echo off\nsetlocal enabledelayedexpansion\nif \"%~1\"==\"\" (\n    for \u002Ff \"tokens=2 delims==\" %%I in ('wmic os get localdatetime \u002Fvalue') do set datetime=%%I\n    set branch_name=worktree-!datetime:~0,8!-!datetime:~8,6!\n) else (\n    set branch_name=%~1\n)\ngit worktree add \"..\u002F%branch_name%\" -b \"%branch_name%\"\nif %errorlevel% equ 0 (\n    cd \"..\u002F%branch_name%\"\n    set ANTHROPIC_AUTH_TOKEN=your-zai-api-key\n    set ANTHROPIC_BASE_URL=https:\u002F\u002Fapi.z.ai\u002Fapi\u002Fanthropic\n    set API_TIMEOUT_MS=3000000\n    claude --model sonnet --permission-mode plan\n)\nendlocal\n```\n\nAfter adding, reload your shell configuration.\n\n**Usage:**\n\n```bash\n# Create worktree with custom name\nzaix feature-auth\n\n# Create worktree with auto-generated timestamp name\nzaix\n```\n\n### Claude Code GitHub Actions\n\nClaude Code can be integrated with GitHub Actions to automate AI-powered workflows. With a simple `@claude` mention in PRs or issues, Claude can analyze code, implement features, fix bugs, and follow project standards defined in `CLAUDE.md`.\n\n**Key capabilities:**\n- Respond to `@claude` mentions in issues and pull requests\n- Create and modify code through pull requests\n- Follow project-specific guidelines from `CLAUDE.md`\n- Execute slash commands like `\u002Freview`\n\n**Reference:** [Claude Code GitHub Actions Documentation](https:\u002F\u002Fcode.claude.com\u002Fdocs\u002Fen\u002Fgithub-actions)\n\n#### Z.AI Integration Example\n\nCreate `.github\u002Fworkflows\u002Fclaude.yml`:\n\n```yaml\nname: Claude Code\n\non:\n  issue_comment:\n    types: [created]\n  pull_request_review_comment:\n    types: [created]\n  issues:\n    types: [opened, assigned]\n  pull_request_review:\n    types: [submitted]\n\njobs:\n  claude:\n    if: |\n      (github.event_name == 'issue_comment' && contains(github.event.comment.body, '@claude')) ||\n      (github.event_name == 'pull_request_review_comment' && contains(github.event.comment.body, '@claude')) ||\n      (github.event_name == 'pull_request_review' && contains(github.event.review.body, '@claude')) ||\n      (github.event_name == 'issues' && (contains(github.event.issue.body, '@claude') || contains(github.event.issue.title, '@claude')))\n    runs-on: ubuntu-latest\n    permissions:\n      contents: write\n      pull-requests: write\n      issues: write\n      id-token: write\n      actions: read\n    steps:\n      - name: Checkout repository\n        uses: actions\u002Fcheckout@v5\n        with:\n          fetch-depth: 1\n\n      - name: Run Claude Code\n        id: claude\n        uses: anthropics\u002Fclaude-code-action@v1\n        env:\n          ANTHROPIC_BASE_URL: https:\u002F\u002Fapi.z.ai\u002Fapi\u002Fanthropic\n          API_TIMEOUT_MS: 3000000\n        with:\n          anthropic_api_key: ${{ secrets.ANTHROPIC_API_KEY }}\n          claude_args: |\n            --model claude-opus\n            --max-turns 100\n```\n\n#### Workflow Explanation\n\n| Component | Purpose |\n|-----------|---------|\n| **Event Triggers** | Listens for `issue_comment`, `pull_request_review_comment`, `issues`, and `pull_request_review` events |\n| **Conditional (`if`)** | Only runs when `@claude` is mentioned in the comment\u002Fissue body or title |\n| **Permissions** | `contents: write` for code changes, `pull-requests: write` for PRs, `issues: write` for issue responses, `actions: read` for CI results |\n| **ANTHROPIC_BASE_URL** | Routes API calls through Z.AI endpoint for higher quotas |\n| **API_TIMEOUT_MS** | Extended timeout (50 minutes) for complex operations |\n| **claude_args** | Uses `claude-opus` model with up to 100 turns for complex tasks |\n\n#### Setup Requirements\n\n1. **Add API key as secret:** Go to repository Settings → Secrets and variables → Actions → Add `ANTHROPIC_API_KEY` with your Z.AI API key\n2. **Create workflow file:** Save the YAML above to `.github\u002Fworkflows\u002Fclaude.yml`\n3. **Usage:** Mention `@claude` in any issue or PR comment to trigger the workflow\n\n## Claude Code Skills\n\nClaude Code now supports [Agent Skills](https:\u002F\u002Fdocs.claude.com\u002Fen\u002Fdocs\u002Fclaude-code\u002Fskills).\n\n### claude-code-guide (Built-in)\n\n- **Purpose**: A natively built-in Claude Code subagent for questions about Claude Code CLI features, hooks, slash commands, MCP servers, settings, IDE integrations, keyboard shortcuts, Claude Agent SDK for building custom agents, and Claude API usage including tool use and Anthropic SDK. No installation required — available automatically in all Claude Code sessions\n\n### consult-zai\n\n* **Purpose**: Dual-AI consultation skill that compares z.ai GLM 4.7 and code-searcher responses for comprehensive code analysis\n* **Location**: `.claude\u002Fskills\u002Fconsult-zai\u002F`\n* **Key Features**:\n  * Invokes both zai-cli and code-searcher agents in parallel for faster response\n  * Enhanced prompts requesting structured output with `file:line` citations\n  * Comparison table showing file paths, line numbers, code snippets, and accuracy\n  * Agreement level indicator (High\u002FPartial\u002FDisagreement) for confidence assessment\n  * Synthesized summary combining best insights from both AI sources\n* **Usage**: `\u002Fconsult-zai \"your code analysis question\"` or invoke via Skill tool\n\n### consult-codex\n\n* **Purpose**: Dual-AI consultation skill that compares OpenAI Codex GPT-5.2 and code-searcher responses for comprehensive code analysis\n* **Location**: `.claude\u002Fskills\u002Fconsult-codex\u002F`\n* **Key Features**:\n  * Invokes both codex-cli and code-searcher agents in parallel for faster response\n  * Enhanced prompts requesting structured output with `file:line` citations\n  * Comparison table showing file paths, line numbers, code snippets, and accuracy\n  * Agreement level indicator (High\u002FPartial\u002FDisagreement) for confidence assessment\n  * Synthesized summary combining best insights from both AI sources\n* **Usage**: `\u002Fconsult-codex \"your code analysis question\"` or invoke via Skill tool\n\n### ai-image-creator\n\n* **Purpose**: Generate PNG images using AI (multiple models via OpenRouter including Gemini, FLUX.2, Riverflow, SeedDream, GPT-5 Image, GPT-5.4 Image 2, proxied through Cloudflare AI Gateway BYOK)\n* **Location**: `.claude\u002Fskills\u002Fai-image-creator\u002F`\n* **Key Features**:\n  * Model selection via keywords: gemini (default), riverflow, flux2, seedream, gpt5, gpt5.4\n  * Supports configurable aspect ratios (1:1, 16:9, 9:16, 3:2, 4:3, etc.) and image sizes (0.5K to 4K)\n  * Multiple providers: OpenRouter (recommended), Google AI Studio, Cloudflare AI Gateway BYOK\n  * Automatic fallback from gateway to direct API\n  * Transparent background generation (`-t`) with green-screen chroma key pipeline\n  * Reference image editing\u002Fstyle transfer (`-r`) for multimodal models (gemini, gpt5, gpt5.4)\n  * Image analysis\u002Fdescription (`--analyze`) — text-only output from multimodal models (gemini, gpt5, gpt5.4); no image generated\n  * Per-project cost tracking (`--costs`) with per-model breakdown\n  * Prompt enhancement with 11 category-specific professional patterns (product hero, social media, marketing banners, etc.)\n  * Composite banners for multi-size logo banners via ImageMagick — no API calls needed\n  * Post-processing support with ImageMagick, sips (macOS), or ffmpeg\n  * Pure Python script with no pip dependencies (requires `uv` runner)\n* **Setup**: Requires API credentials and optional Cloudflare AI Gateway configuration. See [setup guide](.claude\u002Fskills\u002Fai-image-creator\u002Freferences\u002Fsetup-guide.md) for detailed instructions\n* **Usage**: `\u002Fai-image-creator` or invoke via Skill tool when user asks to generate images, create PNGs, make visual assets, or describe\u002Fanalyze existing images\n\n### session-metrics\n\n* **Purpose**: Parse Claude Code's JSONL conversation logs into a per-turn breakdown of token usage, cache efficiency, cost, and activity patterns — useful for understanding your own usage, spotting cache issues, and debugging rate-limit \u002F weekly-session-cap questions\n* **Location**: `.claude\u002Fskills\u002Fsession-metrics\u002F`\n* **Key Features**:\n  * Zero network at runtime — purely local JSONL parsing, no telemetry\n  * Stdlib-only Python (runs via `uv run python`)\n  * Multi-format export: text, JSON, CSV, Markdown, HTML (2-page dashboard + detail split by default, `--single-page` to override)\n  * 5-hour session blocks with trailing 7\u002F14\u002F30-day counters — closest signal available for tracking toward the weekly session cap\n  * Weekly roll-up: trailing 7 days vs prior 7 days with percentage deltas (cost, turns, prompts, blocks, cache hit ratio)\n  * Session duration + burn rate (tokens\u002Fmin, $\u002Fmin) per session\n  * Hour-of-day 24-bucket bars + weekday × hour punchcard in your chosen timezone\n  * Pluggable chart library via `--chart-lib {highcharts,uplot,chartjs,none}` — Highcharts default (vendored + SHA-256 verified, non-commercial licence); uPlot and Chart.js are MIT alternatives; `none` emits a JS-free detail page\n  * Input validation and path-containment checks so `--session` \u002F `--slug` \u002F env vars can't escape `~\u002F.claude\u002Fprojects\u002F`\n  * Instance-wide dashboard (v1.14.0+) via `--all-projects` — aggregates every project under `~\u002F.claude\u002Fprojects\u002F` into a single dated export folder (`exports\u002Fsession-metrics\u002Finstance\u002FYYYY-MM-DD-HHMMSS\u002F`) with per-project HTML drilldowns, daily-cost timeline chart, and a projects breakdown table sorted by cost descending. Add `--no-project-drilldown` for a flat fast-path index; `--projects-dir \u002Fpath` to target a secondary Claude Code install (also honours `CLAUDE_PROJECTS_DIR`)\n* **Usage**: Auto-triggers when you ask questions like \"how much has this session cost?\", \"show me token usage\", or \"session summary\". Manual: `uv run python .claude\u002Fskills\u002Fsession-metrics\u002Fscripts\u002Fsession-metrics.py`. Install via marketplace: `\u002Fplugin marketplace add centminmod\u002Fclaude-plugins` then `\u002Fplugin install session-metrics@centminmod`\n\n## Claude Code Hooks\n\nThe Claude Code hook is for `STOP` which uses Terminal-Notifier to show macOS desktop notifications whenever Claude Code stops and finishes it's response https:\u002F\u002Fgithub.com\u002Fcentminmod\u002Fterminal-notifier-setup.\n\n## Claude Code Subagents\n\nClaude Code subagents are specialized tools designed to handle complex, multi-step tasks autonomously. A key benefit of Claude Code subagents is that uses its own context window separate from the main conversation and can use it's own custom prompt. Learn more about [subagents in the official documentation](https:\u002F\u002Fdocs.anthropic.com\u002Fen\u002Fdocs\u002Fclaude-code\u002Fsub-agents).\n\n### memory-bank-synchronizer\n\n- **Purpose**: Synchronizes memory bank documentation with actual codebase state, ensuring architectural patterns in memory files match implementation reality\n- **Location**: `.claude\u002Fagents\u002Fmemory-bank-synchronizer.md`\n- **Key Responsibilities**:\n  - Pattern documentation synchronization\n  - Architecture decision updates  \n  - Technical specification alignment\n  - Implementation status tracking\n  - Code example freshness validation\n  - Cross-reference validation\n- **Usage**: Proactively maintains consistency between CLAUDE-*.md files and source code to ensure documentation remains accurate and trustworthy\n\n### code-searcher\n\n- **Purpose**: A specialized agent for efficiently searching the codebase, finding relevant files, and summarizing code. Supports both standard detailed analysis and optional [Chain of Draft (CoD)](https:\u002F\u002Fgithub.com\u002Fcentminmod\u002For-cli\u002Fblob\u002Fmaster\u002Fexamples\u002Fexample-code-inspection-prompts3.md) ultra-concise mode when explicitly requested for 80% token reduction\n- **Location**: `.claude\u002Fagents\u002Fcode-searcher.md`\n- **Key Responsibilities**:\n  - Efficient codebase navigation and search\n  - Function and class location\n  - Code pattern identification\n  - Bug source location assistance\n  - Feature implementation analysis\n  - Integration point discovery\n  - Chain of Draft (CoD) mode for ultra-concise reasoning with minimal tokens\n- **Usage**: Use when you need to locate specific functions, classes, or logic within the codebase. Request \"use CoD\", \"chain of draft\", or \"draft mode\" for ultra-concise responses with ~80% fewer tokens\n  - **Standard mode**: \"Find the payment processing code\" → Full detailed analysis\n  - **CoD mode**: \"Find the payment processing code using CoD\" → \"Payment→glob:*payment*→found:payment.service.ts:45\"\n\n### get-current-datetime\n\n- **Purpose**: Simple DateTime utility for accurate Brisbane, Australia (GMT+10) timezone values. Executes bash date commands and returns only the raw output without formatting or explanations\n- **Location**: `.claude\u002Fagents\u002Fget-current-datetime.md`\n- **Key Responsibilities**:\n  - Execute `TZ='Australia\u002FBrisbane' date` commands\n  - Provide accurate Brisbane timezone timestamps\n  - Support multiple format options (default, filename, readable, ISO)\n  - Eliminate timezone confusion and month errors\n  - Return raw command output without additional processing\n- **Usage**: Use when creating files with timestamps, generating reports with dates, or needing accurate Australian timezone values for any purpose\n\n### ux-design-expert\n\n- **Purpose**: Comprehensive UX\u002FUI design guidance specialist combining user experience optimization, premium interface design, and scalable design systems with Tailwind CSS and Highcharts data visualization\n- **Location**: `.claude\u002Fagents\u002Fux-design-expert.md`\n- **Key Responsibilities**:\n  - UX flow optimization and friction reduction\n  - Premium UI design with sophisticated visual hierarchies\n  - Scalable design systems architecture using Tailwind CSS\n  - Data visualization strategy with Highcharts implementations\n  - Accessibility compliance and performance optimization\n  - Component library design with atomic methodology\n- **Usage**: Use for dashboard UX improvements, premium component libraries, complex user flow optimization, design system creation, or any comprehensive UX\u002FUI design guidance needs\n\n### zai-cli\n\n* **Purpose**: CLI wrapper agent for executing z.ai GLM 4.7 model queries\n* **Location**: `.claude\u002Fagents\u002Fzai-cli.md`\n* **Key Features**:\n  * Executes z.ai CLI with JSON output format\n  * Uses haiku model for minimal overhead\n  * Returns raw output for parent skill to process\n* **Usage**: Used internally by consult-zai skill; not typically invoked directly\n\n### codex-cli\n\n* **Purpose**: CLI wrapper agent for executing OpenAI Codex GPT-5.2 queries\n* **Location**: `.claude\u002Fagents\u002Fcodex-cli.md`\n* **Key Features**:\n  * Executes Codex CLI in readonly mode with JSON output\n  * Uses haiku model for minimal overhead\n  * Returns raw output for parent skill to process\n* **Usage**: Used internally by consult-codex skill; not typically invoked directly\n\n## Claude Code Slash Commands\n\n### `\u002Fanthropic` Commands\n\n- **`\u002Fapply-thinking-to`** - Expert prompt engineering specialist that applies Anthropic's extended thinking patterns to enhance prompts with advanced reasoning frameworks\n  - Transforms prompts using progressive reasoning structure (open-ended → systematic)\n  - Applies sequential analytical frameworks and systematic verification with test cases\n  - Includes constraint optimization, bias detection, and extended thinking budget management\n  - Usage: `\u002Fapply-thinking-to @\u002Fpath\u002Fto\u002Fprompt-file.md`\n\n- **`\u002Fconvert-to-todowrite-tasklist-prompt`** - Converts complex, context-heavy prompts into efficient TodoWrite tasklist-based methods with parallel subagent execution\n  - Achieves 60-70% speed improvements through parallel processing\n  - Transforms verbose workflows into specialized task delegation\n  - Prevents context overflow through strategic file selection (max 5 files per task)\n  - Usage: `\u002Fconvert-to-todowrite-tasklist-prompt @\u002Fpath\u002Fto\u002Foriginal-slash-command.md`\n\n- **`\u002Fupdate-memory-bank`** - Simple command to update CLAUDE.md and memory bank files\n  - Usage: `\u002Fupdate-memory-bank`\n\n### `\u002Fccusage` Commands\n\n- **`\u002Fccusage-daily`** - Generates comprehensive Claude Code usage cost analysis and statistics\n  - Runs `ccusage daily` command and parses output into structured markdown\n  - Provides executive summary with total costs, peak usage days, and cache efficiency\n  - Creates detailed tables showing daily costs, token usage, and model statistics\n  - Includes usage insights, recommendations, and cost management analysis\n  - Usage: `\u002Fccusage-daily`\n\n### `\u002Fcleanup` Commands\n\n- **`\u002Fcleanup-context`** - Memory bank optimization specialist for reducing token usage in documentation\n  - Removes duplicate content and eliminates obsolete files\n  - Consolidates overlapping documentation while preserving essential information\n  - Implements archive strategies for historical documentation\n  - Achieves 15-25% token reduction through systematic optimization\n  - Usage: `\u002Fcleanup-context`\n\n### `\u002Fdocumentation` Commands\n\n- **`\u002Fcreate-readme-section`** - Generate specific sections for README files with professional formatting\n  - Creates well-structured sections like Installation, Usage, API Reference, Contributing, etc.\n  - Follows markdown best practices with proper headings, code blocks, and formatting\n  - Analyzes project context to provide relevant content\n  - Matches existing README style and tone\n  - Usage: `\u002Fcreate-readme-section \"Create an installation section for my Python project\"`\n\n- **`\u002Fcreate-release-note`** - Generate comprehensive release documentation from recent commits with dual output formats\n  - Interactive workflow with two modes: by commit count or by commit hash range (last 24\u002F48\u002F72 hours)\n  - Produces customer-facing release note (value-focused, no technical jargon) and technical engineering note (SHA references, file paths)\n  - Comprehensive commit analysis with grouping by subsystem and traceability to specific SHAs\n  - Supports direct arguments for quick generation or interactive selection for precise control\n  - Usage: `\u002Fcreate-release-note` (interactive), `\u002Fcreate-release-note 20` (last 20 commits), or select commit hash after viewing recent commits\n\n### `\u002Fsecurity` Commands\n\n- **`\u002Fsecurity-audit`** - Perform comprehensive security audit of the codebase\n  - Identifies potential vulnerabilities using OWASP guidelines\n  - Checks authentication, input validation, data protection, and API security\n  - Categorizes issues by severity (Critical, High, Medium, Low)\n  - Provides specific remediation steps with code examples\n  - Usage: `\u002Fsecurity-audit`\n\n- **`\u002Fcheck-best-practices`** - Analyze code against language-specific best practices\n  - Detects languages and frameworks to apply relevant standards\n  - Checks naming conventions, code organization, error handling, and performance\n  - Provides actionable feedback with before\u002Fafter code examples\n  - Prioritizes impactful improvements over nitpicks\n  - Usage: `\u002Fcheck-best-practices`\n\n- **`\u002Fsecure-prompts`** - Enterprise-grade security analyzer for detecting prompt injection attacks and malicious instructions\n  - Detects prompt injection attacks, hidden content, and malicious instructions using advanced AI-specific detection patterns\n  - Provides comprehensive threat analysis with automated timestamped report generation\n  - Saves reports to `reports\u002Fsecure-prompts\u002F` directory for audit trails\n  - Analyzes both file content and direct text input for security threats\n  - Usage: `\u002Fsecure-prompts @suspicious_file.txt` or `\u002Fsecure-prompts \"content to analyze\"`\n  - Example prompt injection prompts at `.claude\u002Fcommands\u002Fsecurity\u002Ftest-examples` that you can run `\u002Fsecure-prompts` against.\n  - Example generated report for `\u002Fsecure-prompts .claude\u002Fcommands\u002Fsecurity\u002Ftest-examples\u002Ftest-encoding-attacks.md` [here](reports\u002Fsecure-prompts\u002Fsecurity-analysis_20250719_072359.md)\n\n### `\u002Farchitecture` Commands\n\n- **`\u002Fexplain-architecture-pattern`** - Identify and explain architectural patterns in the codebase\n  - Analyzes project structure and identifies design patterns\n  - Explains rationale behind architectural decisions\n  - Provides visual representations with diagrams\n  - Shows concrete implementation examples\n  - Usage: `\u002Fexplain-architecture-pattern`\n\n### `\u002Fpromptengineering` Commands\n\n- **`\u002Fconvert-to-test-driven-prompt`** - Transform requests into Test-Driven Development style prompts\n  - Defines explicit test cases with Given\u002FWhen\u002FThen format\n  - Includes success criteria and edge cases\n  - Structures prompts for red-green-refactor cycle\n  - Creates measurable, specific test scenarios\n  - Usage: `\u002Fconvert-to-test-driven-prompt \"Add user authentication feature\"`\n\n- **`\u002Fbatch-operations-prompt`** - Optimize prompts for multiple file operations and parallel processing\n  - Identifies parallelizable tasks to maximize efficiency\n  - Groups operations by conflict potential\n  - Integrates with TodoWrite for task management\n  - Includes validation steps between batch operations\n  - Usage: `\u002Fbatch-operations-prompt \"Update all API calls to use new auth header\"`\n\n### `\u002Frefactor` Commands\n\n- **`\u002Frefactor-code`** - Analysis-only refactoring specialist that creates comprehensive refactoring plans without modifying code\n  - Analyzes code complexity, test coverage, and architectural patterns\n  - Identifies safe extraction points and refactoring opportunities\n  - Creates detailed step-by-step refactoring plans with risk assessment\n  - Generates timestamped reports in `reports\u002Frefactor\u002F` directory\n  - Focuses on safety, incremental progress, and maintainability\n  - Usage: `\u002Frefactor-code`\n\n## Claude Code Plan Weekly Rate Limits\n\nIf you are using Claude monthly subscription plans for Claude Code, new weekly rate limits will apply from August 28, 2025 in addition to max 50x 5hr session limits per month:\n\n| Plan               | Sonnet 4 (hrs\u002Fweek) | Opus 4 (hrs\u002Fweek) |\n|--------------------|---------------------|-------------------|\n| Pro                | 40–80               | –                 |\n| Max ($100 \u002Fmo)     | 140–280             | 15–35             |\n| Max ($200 \u002Fmo)     | 240–480             | 24–40             |\n\n## Claude Code settings\n\n> Configure Claude Code with global and project-level settings, and environment variables.\n\nClaude Code offers a variety of settings to configure its behavior to meet your needs. You can configure Claude Code by running the `\u002Fconfig` command when using the interactive REPL.\n\n### Settings files\n\nThe `settings.json` file is our official mechanism for configuring Claude\nCode through hierarchical settings:\n\n* **User settings** are defined in `~\u002F.claude\u002Fsettings.json` and apply to all\n  projects.\n* **Project settings** are saved in your project directory:\n  * `.claude\u002Fsettings.json` for settings that are checked into source control and shared with your team\n  * `.claude\u002Fsettings.local.json` for settings that are not checked in, useful for personal preferences and experimentation. Claude Code will configure git to ignore `.claude\u002Fsettings.local.json` when it is created.\n\n#### Available settings\n\n`settings.json` supports a number of options:\n\n| Key                   | Description                                                                                                                                                                                                    | Example                         |\n| :-------------------- | :------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | :------------------------------ |\n| `apiKeyHelper`        | Custom script, to be executed in `\u002Fbin\u002Fsh`, to generate an auth value. This value will generally be sent as `X-Api-Key`, `Authorization: Bearer`, and `Proxy-Authorization: Bearer` headers for model requests | `\u002Fbin\u002Fgenerate_temp_api_key.sh` |\n| `cleanupPeriodDays`   | How long to locally retain chat transcripts (default: 30 days)                                                                                                                                                 | `20`                            |\n| `env`                 | Environment variables that will be applied to every session                                                                                                                                                    | `{\"FOO\": \"bar\"}`                |\n| `includeCoAuthoredBy` | Whether to include the `co-authored-by Claude` byline in git commits and pull requests (default: `true`)                                                                                                       | `false`                         |\n| `permissions`         | See table below for structure of permissions.                                                                                                                                                                  |                                 |\n\n#### Permission settings\n\n| Keys                           | Description                                                                                                                                        | Example                          |\n| :----------------------------- | :------------------------------------------------------------------------------------------------------------------------------------------------- | :------------------------------- |\n| `allow`                        | Array of [permission rules](\u002Fen\u002Fdocs\u002Fclaude-code\u002Fiam#configuring-permissions) to allow tool use                                                    | `[ \"Bash(git diff:*)\" ]`         |\n| `deny`                         | Array of [permission rules](\u002Fen\u002Fdocs\u002Fclaude-code\u002Fiam#configuring-permissions) to deny tool use                                                     | `[ \"WebFetch\", \"Bash(curl:*)\" ]` |\n| `additionalDirectories`        | Additional [working directories](iam#working-directories) that Claude has access to                                                                | `[ \"..\u002Fdocs\u002F\" ]`                 |\n| `defaultMode`                  | Default [permission mode](iam#permission-modes) when opening Claude Code                                                                           | `\"acceptEdits\"`                  |\n| `disableBypassPermissionsMode` | Set to `\"disable\"` to prevent `bypassPermissions` mode from being activated. See [managed policy settings](iam#enterprise-managed-policy-settings) | `\"disable\"`                      |\n\n#### Settings precedence\n\nSettings are applied in order of precedence:\n\n1. Enterprise policies (see [IAM documentation](\u002Fen\u002Fdocs\u002Fclaude-code\u002Fiam#enterprise-managed-policy-settings))\n2. Command line arguments\n3. Local project settings\n4. Shared project settings\n5. User settings\n\n### Environment variables\n\nClaude Code supports the following environment variables to control its behavior:\n\n\u003CNote>\n  All environment variables can also be configured in [`settings.json`](#available-settings). This is useful as a way to automatically set environment variables for each session, or to roll out a set of environment variables for your whole team or organization.\n\u003C\u002FNote>\n\n| Variable                                   | Purpose                                                                                                                                |\n| :----------------------------------------- | :------------------------------------------------------------------------------------------------------------------------------------- |\n| `ANTHROPIC_API_KEY`                        | API key sent as `X-Api-Key` header, typically for the Claude SDK (for interactive usage, run `\u002Flogin`)                                 |\n| `ANTHROPIC_AUTH_TOKEN`                     | Custom value for the `Authorization` and `Proxy-Authorization` headers (the value you set here will be prefixed with `Bearer `)        |\n| `ANTHROPIC_CUSTOM_HEADERS`                 | Custom headers you want to add to the request (in `Name: Value` format)                                                                |\n| `ANTHROPIC_MODEL`                          | Name of custom model to use (see [Model Configuration](\u002Fen\u002Fdocs\u002Fclaude-code\u002Fbedrock-vertex-proxies#model-configuration))               |\n| `ANTHROPIC_SMALL_FAST_MODEL`               | Name of [Haiku-class model for background tasks](\u002Fen\u002Fdocs\u002Fclaude-code\u002Fcosts)                                                           |\n| `ANTHROPIC_SMALL_FAST_MODEL_AWS_REGION`    | Override AWS region for the small\u002Ffast model when using Bedrock                                                                        |\n| `BASH_DEFAULT_TIMEOUT_MS`                  | Default timeout for long-running bash commands                                                                                         |\n| `BASH_MAX_TIMEOUT_MS`                      | Maximum timeout the model can set for long-running bash commands                                                                       |\n| `BASH_MAX_OUTPUT_LENGTH`                   | Maximum number of characters in bash outputs before they are middle-truncated                                                          |\n| `CLAUDE_BASH_MAINTAIN_PROJECT_WORKING_DIR` | Return to the original working directory after each Bash command                                                                       |\n| `CLAUDE_CODE_API_KEY_HELPER_TTL_MS`        | Interval in milliseconds at which credentials should be refreshed (when using `apiKeyHelper`)                                          |\n| `CLAUDE_CODE_IDE_SKIP_AUTO_INSTALL`        | Skip auto-installation of IDE extensions (defaults to false)                                                                           |\n| `CLAUDE_CODE_MAX_OUTPUT_TOKENS`            | Set the maximum number of output tokens for most requests                                                                              |\n| `CLAUDE_CODE_USE_BEDROCK`                  | Use [Bedrock](\u002Fen\u002Fdocs\u002Fclaude-code\u002Famazon-bedrock)                                                                                     |\n| `CLAUDE_CODE_USE_VERTEX`                   | Use [Vertex](\u002Fen\u002Fdocs\u002Fclaude-code\u002Fgoogle-vertex-ai)                                                                                    |\n| `CLAUDE_CODE_SKIP_BEDROCK_AUTH`            | Skip AWS authentication for Bedrock (e.g. when using an LLM gateway)                                                                   |\n| `CLAUDE_CODE_SKIP_VERTEX_AUTH`             | Skip Google authentication for Vertex (e.g. when using an LLM gateway)                                                                 |\n| `CLAUDE_CODE_DISABLE_NONESSENTIAL_TRAFFIC` | Equivalent of setting `DISABLE_AUTOUPDATER`, `DISABLE_BUG_COMMAND`, `DISABLE_ERROR_REPORTING`, and `DISABLE_TELEMETRY`                 |\n| `DISABLE_AUTOUPDATER`                      | Set to `1` to disable automatic updates. This takes precedence over the `autoUpdates` configuration setting.                           |\n| `DISABLE_BUG_COMMAND`                      | Set to `1` to disable the `\u002Fbug` command                                                                                               |\n| `DISABLE_COST_WARNINGS`                    | Set to `1` to disable cost warning messages                                                                                            |\n| `DISABLE_ERROR_REPORTING`                  | Set to `1` to opt out of Sentry error reporting                                                                                        |\n| `DISABLE_NON_ESSENTIAL_MODEL_CALLS`        | Set to `1` to disable model calls for non-critical paths like flavor text                                                              |\n| `DISABLE_TELEMETRY`                        | Set to `1` to opt out of Statsig telemetry (note that Statsig events do not include user data like code, file paths, or bash commands) |\n| `HTTP_PROXY`                               | Specify HTTP proxy server for network connections                                                                                      |\n| `HTTPS_PROXY`                              | Specify HTTPS proxy server for network connections                                                                                     |\n| `MAX_THINKING_TOKENS`                      | Force a thinking for the model budget                                                                                                  |\n| `MCP_TIMEOUT`                              | Timeout in milliseconds for MCP server startup                                                                                         |\n| `MCP_TOOL_TIMEOUT`                         | Timeout in milliseconds for MCP tool execution                                                                                         |\n| `MAX_MCP_OUTPUT_TOKENS`                    | Maximum number of tokens allowed in MCP tool responses (default: 25000)                                                                |\n| `VERTEX_REGION_CLAUDE_3_5_HAIKU`           | Override region for Claude 3.5 Haiku when using Vertex AI                                                                              |\n| `VERTEX_REGION_CLAUDE_3_5_SONNET`          | Override region for Claude 3.5 Sonnet when using Vertex AI                                                                             |\n| `VERTEX_REGION_CLAUDE_3_7_SONNET`          | Override region for Claude 3.7 Sonnet when using Vertex AI                                                                             |\n| `VERTEX_REGION_CLAUDE_4_0_OPUS`            | Override region for Claude 4.0 Opus when using Vertex AI                                                                               |\n| `VERTEX_REGION_CLAUDE_4_0_SONNET`          | Override region for Claude 4.0 Sonnet when using Vertex AI                                                                             |\n\n### Configuration options\n\nWe are in the process of migrating global configuration to `settings.json`.\n\n`claude config` will be deprecated in place of [settings.json](#settings-files)\n\nTo manage your configurations, use the following commands:\n\n* List settings: `claude config list`\n* See a setting: `claude config get \u003Ckey>`\n* Change a setting: `claude config set \u003Ckey> \u003Cvalue>`\n* Push to a setting (for lists): `claude config add \u003Ckey> \u003Cvalue>`\n* Remove from a setting (for lists): `claude config remove \u003Ckey> \u003Cvalue>`\n\nBy default `config` changes your project configuration. To manage your global configuration, use the `--global` (or `-g`) flag.\n\n#### Global configuration\n\nTo set a global configuration, use `claude config set -g \u003Cke","该项目提供了一个Claude Code的启动配置模板和记忆库系统，旨在帮助开发者更有效地利用Claude AI进行开发。核心功能包括基于官方Anthropic最佳实践设计的三种不同风格的CLAUDE.md模板，支持双内存架构、渐进式披露以及独立的行为规则，并为现有用户提供迁移指南以适应新模板。项目使用Python编写，适合需要长时间对话保持上下文连贯性的AI应用开发场景，如复杂问题咨询、多轮对话系统构建等。","2026-06-11 03:42:24","high_star"]