All AI tools

claude-code-local is an open-source project that lets Claude Code run entirely locally on an Apple Silicon Mac. It responds to Claude Code directly in the Anthropic Messages API format through a Python server based on Apple MLX (proxy/server.py), requiring no cloud model and no API Key; code and data never leave the computer, and it supports offline and air-gapped environments. The project provides installation and startup scripts, tool-call parsing tests, a native terminal coding agent with cache optimization, session continuation tools, and a desktop launcher, and automatically selects a runnable local model based on Mac memory. It is suitable for privacy-sensitive workflows such as NDA, legal, and medical work.

claude-code-open

Open-source AI coding platform based on Claude Code CLI with full MCP protocol support. Features a browser-based Web IDE with Monaco editor, 37+ built-in tools, Blueprint multi-agent orchestration system, and scheduled task daemon. Supports MCP server configuration and integration for extending AI c

claude-code-pro-pack

claude-code-pro-pack is a collection of behavioral rules and skill templates for AI coding agents, suitable for Claude Code, Codex, Cursor, and Hermes Agent. At its core is the CLAUDE.md or AGENTS.md twelve-rule behavior file, which claims to reduce AI coding errors from about 40% to about 3%, with a total size of about 700 tokens, and it takes effect simply by placing it in the project root directory. The pack is built on real agent failure patterns from more than 30 codebases, supplementing agent orchestration issues not covered by the original 4 rules, such as token spirals, silent partial failures, dual-mode contamination, and duplicate function drift. The project uses the MIT license and includes the cc-audit auditing tool and example skill templates.

claude-code-router

A powerful tool to route Claude Code requests to different models and customize any request. Supports multi-provider routing, request/response transformation, dynamic model switching, CLI model management, and GitHub Actions integration.

claude-codex-bridge

Run Codex agents from Claude Code through MCP and Channels, with live progress notifications, mid-task steering and session continuation.

claude-concilium

Multi-agent consultation for Claude Code: MCP servers wrapping Codex, Gemini and Qwen CLIs for parallel code review and problem-solving with fallback chains.

claude-consensus

claw-consensus (formerly claude-consensus) is a reference implementation of a multi-machine coordination layer for running AI agents on multiple machines and avoiding state drift. Through a consensus protocol (propose/counter/accept/commit), a dual-track message bus, ACK acknowledgment discipline, and self-healing synchronization mechanisms, it enables agents on different computers to negotiate decisions, commit results, and verify each other, waking humans only when money or irreversible operations are involved. The project is implemented purely with the Python standard library, requires no additional packages, servers, or databases, and is free and open source under the MIT license.

claude-critical-rules-mcp

Enforces critical rules for Claude to prevent documented failure patterns, with a compliance verification checklist and rules summary.

claude-debugs-for-you

VS Code extension and MCP server for language-agnostic automatic debugging via breakpoints and expression evaluation.

claude-desktop-extension-bear-notes

AI productivity

Search, read, create and update Bear notes locally.

claude-engram

Local persistent memory for Claude Code: tracks mistakes and decisions via hooks, searches session history, detects loops and warns before edits.

claude-faf-mcp

Persistent project context in the .faf (Foundational AI-context Format), with project DNA and Podium scoring.

claude-find

Retrieve memory from across your Claude Code sessions on demand.

claude-flow

Agent orchestration platform with a Queen + Workers architecture — manages Claude, GPT, and Gemini simultaneously via MCP.

claude-handoff

Claude Code session handoffs: turns any transcript, even crashed ones, into a handoff file and distills project history into a cited memory brief.

claude-lens

claude-pace (repository name claude-lens) is a lightweight Claude Code status bar and rate limit tracking tool that displays 5-hour and 7-day quota usage in real time in the terminal, along with a reset countdown. Its biggest feature is "pace-aware" quota monitoring: it not only tells you how much you have used, but also compares the current consumption rate with the remaining time in the window, using ⇡ or ⇣ to show whether you are overspending. The tool is written in pure Bash + jq, is a single file, has zero npm dependencies, and can be installed via the Claude Code plugin, npx, or a script.

claude-mcp-bridge

Wraps the Claude Code CLI as MCP tools for queries, search, structured output and sessions, with budget caps and granular tool sandboxing.

Claude-MCP-Read-Email-Attachments

AI office

Read Outlook email attachments via Microsoft Graph, parsing PDF, Word (including embedded images), Excel and text files into structured content.

claude-mermaid

Render Mermaid diagrams with live reload, themes and SVG, PNG or PDF export.

claude-pet

Desktop companion for Claude Code that reacts to hook events with animated emotions, maintains a local graph memory per project, promotes repeated patterns into skill files, and monitors GitHub repository activity. macOS/Windows/Linux, 100% local, MIT.

claude-plugin

🎖️ ☁️ - Product management: link customer feedback to features, releases and objectives, with roadmap-drift detection and funnel, retention, NPS and NRR analytics.

claude-plugin

🎖️ 🐍 ☁️ - Scan and fix a domain's email authentication (SPF, DMARC, DKIM, MX), blacklist status and domain/SSL expiry, with deterministic, validated fix records.

claude-plugins-community

Community plugin marketplace for Claude Cowork and Claude Code. Read-only mirror — submit plugins at clau.de/plugin-directory-submission.

claude-plugins-official

Official, Anthropic-managed directory of high quality Claude Code Plugins.

claude-prism

AI writing

ClaudePrism is a local-first scientific writing workbench, powered by Claude, integrating LaTeX, Python, and 100+ scientific skills, and runs on the desktop. Documents are stored on the local disk and compiled offline; when using AI features, prompts and file contents are sent to the Anthropic API for inference. It supports quickly creating projects from templates, embedded Tectonic offline LaTeX compilation, Python virtual environments and code execution, a Claude chat assistant, local Git history snapshots and diff comparison, real-time PDF preview with SyncTeX jumping, Zotero reference management, and custom slash commands, making it suitable for academic writing scenarios such as papers, theses, presentations, and posters.

claude-scientific-skills

AI search

Scientific Agent Skills (formerly Claude Scientific Skills) is an open-source scientific research skills library maintained by K-Dense, containing 177 validated scientific research skills, covering fields such as biology, chemistry, medicine, drug discovery, physics, engineering, earth science, data analysis, and scientific communication. It follows the open Agent Skills standard, enabling AI agents such as Cursor, Claude Code, Codex, and Google Antigravity to gain professional scientific research workflow capabilities, and connects to more than 100 scientific databases.

claude-session-relay

Self-hosted relay where coding-agent sessions on different machines share a team board, message each other and check for git conflicts, with isolated teams, private sessio

claude-skills

14-category autonomous product-building OS with 18 specialized agents that turn one-line prompts into deployed products on Cloudflare Workers.

claude-skills

14-category autonomous product-building OS bundling 18 specialized agents (architect, test-writer, deploy-verifier, security-reviewer, visual-qa, performance-profiler, accessibility-auditor, content-writer, completeness-checker, dependency-auditor, migration-agent, incident-responder, computer-use-o

claude-skills-mcp

Search and use Claude Agent Skills from any model or client as if natively supported.

claude-snapshot

claude-snapshot is a portable configuration snapshot tool for Claude Code that exports your configuration, plugins, hooks, and global instructions as a .tar.gz snapshot file and quickly applies it on another machine. It supports scenarios such as multi-machine synchronization, backup before system reinstallation, safe rollback, and team onboarding. By design, it adheres to a minimal footprint (no daemon, no background listening), allowlist collection, never connecting to the network, diff before overwrite, and cross-platform first (Node.js implementation). It requires Node.js 18+, can be used as a Claude Code plugin, and can also be run via npx without installation. Current version v0.2.

claude-soul

Learning engine for Claude Code: extracts signals from sessions, runs periodic reflections and promotes or retires behavioral frameworks based on evidence.

claude-squad

Terminal app for running multiple Claude Code and Codex sessions simultaneously, each in its own workspace.

claude-terminal-mcp

Terminal, filesystem and background-job tools for Linux and macOS.

claude-to-chatgpt