* [opencodex v2.83.0](https://github.com/lidge-jun/opencodex) – Lightweight local proxy that translates Codex Responses API into multiple LLM providers with full support for streaming, tools, reasoning tokens, and images.

* [Skills Hub v1.0.0](https://github.com/qufei1993/skills-hub) – Cross-platform Tauri and React app for installing, organizing, updating, and syncing Agent Skills across multiple AI coding tools’ skills directories.

* [Clawd on Desk v1.3.0](https://github.com/rullerzhou-afk/clawd-on-desk) – Desktop pet reacting to AI coding agent sessions in real time with animations for thinking, typing, juggling, permission prompts, and sleep.

* [Token Monitor v0.68.0](https://github.com/Javis603/token-monitor) – Desktop widget that monitors live token usage and AI tool limits across multiple AI coding tools with real-time multi-device sync and usage history breakdowns.

* [Marketing Skills for Claude Code v2.11.19](https://github.com/coreyhaines31/marketingskills) – Collection of AI agent skills for conversion optimization, copywriting, SEO, analytics, and growth engineering.

* [CLI Proxy API v8.0.19](https://github.com/router-for-me/CLIProxyAPI) – Proxy server providing OpenAI, Gemini, Claude, Codex, and Qwen compatible API interfaces for CLI models with multi-account support.

* [PenEcho v1.4.0](https://github.com/penecho/penecho) – A shared canvas that reads handwriting and diagrams with spatial context and generates editable AI answers beside the marks.

* [AgentSys v6.1.0](https://github.com/agent-sh/agentsys) – Modular runtime and orchestration system for composing AI agents, plugins, and skills into structured development pipelines.

* [fauxnix v2.0.0](https://github.com/20000419/fauxnix) – Bash-to-PowerShell translation layer for Windows that lets AI agents run familiar Linux commands with GNU-style output and deterministic execution via an MCP server and CLI.

* [Codync v2.14.0](https://github.com/leepokai/Codync) – Codync is a Rust-hosted, iPhone-to-desktop companion that turns multiple coding agents into persistent chat bots with approvals, notifications, and remote screen control.

* [Oil UI v0.21.0](https://github.com/oil-oil/oil-ui) – Screenshot-driven AI UI agent for exploring multiple design directions, comparing them side by side, and refining results.

* [Cherry Studio v2.1.6](https://github.com/CherryHQ/cherry-studio) – Desktop client supporting multiple LLM providers, available on Windows, Mac, and Linux.

* [ChatGPT Web for Codex v6.1.8](https://github.com/miuuyy/codex-chatgpt-web) – ChatGPT Web bridge that runs selectable ChatGPT tiers as native Codex models with full task context, images, streaming, and MCP tool access.

* [OpenBot v0.36.0](https://github.com/nightly-labs/openbot) – Local-first desktop workspace where persistent AI agents get per-agent queues, file transfers, embedded browser control, and agent-to-agent messaging.

* [OpenMausBot v0.1.107](https://github.com/milind-soni/OpenMausBot) – Local-first chat app that runs multiple AI agents with separate models, memory, and connected apps on your own machine.

* [Clodex v5.122.0](https://github.com/avirtual/clodex) – Visual manager for fleets of coding agents, showing real terminal sessions with live context, messaging, cost telemetry, and remote coordination across Mac and Linux.

* [Paseo v0.11.2](https://github.com/getpaseo/paseo) – Run and manage coding agents on your machines from phone, desktop, web, and CLI.

* [ReevesAgents v1.8.0](https://github.com/mertkayacs/reevesagents) – Local tmux-first workspace that lets one AI coding agent spawn, steer, and supervise other agent CLIs via MCP with shared TUI, Web UI, and CLI control.

* [AIClient2APIA2 v3.5.7](https://github.com/justlovemaki/AIClient2API) – Proxy that unifies client-only LLM APIs into an OpenAI-compatible local interface and simulates model requests.

* [Sub2API v0.2.15](https://github.com/Wei-Shaw/sub2api) – AI API gateway that distributes subscription quotas, manages keys, billing, load balancing, and request forwarding.

* [lnwjud v5.8.0](https://github.com/engasnm111/lnwjud) – Windows-first local AI-agent runtime and MCP gateway that exposes trusted capabilities through MCP for files, Git, processes, Windows automation, WSL, browser control, indexing, and observability.

* [codex-cli-nix v0.162.0](https://github.com/sadjow/codex-cli-nix) – Nix flake providing an always-up-to-date, native Rust OpenAI Codex CLI with hourly updates and multi-platform binary cache.

* [Solana Development Skill v2.6.0](https://github.com/solana-foundation/solana-dev-skill) – A comprehensive agent skill for modern Solana development, covering SDK, UI, legacy interop, programs, testing, codegen, security, and toolchain best practices.

* [EasyRead v1.4.0](https://github.com/Edwardxlai/easyread) – EasyRead converts and enriches PDF content with KaTeX, AI summaries, and structured export for easier reading and review.

* [oc-codex-multi-auth v6.28.0](https://github.com/ndycode/oc-codex-multi-auth) – OpenCode plugin enabling ChatGPT Plus/Pro OAuth access, GPT-5/Codex model presets, multi-account rotation with failover, and guided setup.

* [AFFiNE MCP Server v3.9.0](https://github.com/DAWNCR0W/affine-mcp-server) – Model Context Protocol server that exposes AFFiNE workspaces, documents, and database workflows to AI assistants over stdio or HTTP.

* [Codex Security npm-v0.2.0](https://github.com/openai/codex-security) – A CLI and TypeScript SDK for finding, validating, and fixing security vulnerabilities in your code.

* [reminal v3.16.0](https://github.com/harshalgajjar/Reminal) – Reminal streams and mirrors a machine’s real windows and full terminal into any browser with end-to-end encryption and no open ports.

* [Codexia v0.54.0](https://github.com/milisp/codexia) – GUI and toolkit for managing Codex CLI projects, sessions, file-tree navigation, prompt notepad, git worktrees, file viewers, and usage analytics.

* [Archify archify-dsh-v1.0.0](https://github.com/tt-a1i/archify) – Tool that generates workflow, sequence, data-flow, and lifecycle architecture diagrams with light/dark themes and multi-format export.

* [META\_KIM v3.4.0](https://github.com/KimYx0207/Meta_Kim) – Governance layer that coordinates AI coding assistants via an 8-stage workflow, contracts, gates, and memory to improve execution and learning.

* [Daintree v0.43.0](https://github.com/daintreehq/daintree) – Daintree is a macro-orchestration app for running many AI coding agent terminals in parallel across git worktrees with context injection and workflow automation.

* [iLab CONJURE v0.9.7](https://github.com/kadevin/ilab-conjure) – iLab CONJURE is a FastAPI-based AI image generation WebUI and CLI with OpenAI-compatible and Codex Responses API support, featuring templates, chips, and local history queues.

* [Kenfold v0.1.0](https://github.com/jeonjw85/kenfold) – Self-hosted, typed memory server that shares agent decisions and context across multiple AI agents via MCP with provenance, history, and poisoning-aware storage.

* [computer-use-linux v0.7.15](https://github.com/agent-sh/computer-use-linux) – Linux desktop control server and CLI for MCP hosts, offering AT-SPI accessibility, compositor-aware window targeting, screenshots, and input.

* [DevCodex v1.20.4](https://github.com/devcodex-labs/devcodex) – Intent-driven AI coding workflow runtime that keeps consistent context, skills, approvals, validation, and handoffs across six AI coding hosts.

* [Codex App Manager v0.5.16](https://github.com/Wangnov/Codex-App-Manager) – Cross-platform desktop installer and updater that applies incremental updates, supports clean uninstalls, and performs China-reachable self-updates for the official Codex app.

* [Orca v1.4.224](https://github.com/stablyai/orca) – Cross-platform AI orchestrator for managing multiple worktrees, running concurrent coding agents, and tracking agent progress.

* [Claudexor v3.25.1](https://github.com/razzant/claudexor) – Local-first control plane that runs multiple AI coding agents via a typed interface, adding quota-aware routing, best-of-N races, and auditable execution evidence.

* [nodeterm v0.4.3](https://github.com/eneskirca/nodeterm) – Node-based terminal and AI session manager that maps persistent tmux sessions onto a draggable infinite canvas with optional Trello-style project boards.

* [Harness Mix v0.4.1](https://github.com/emo-xiaoyu/harness-mix) – Codex Desktop-style coding harness that provides a host runtime and protocol core across Windows, macOS, and Linux.

* [tmux-ide v2.9.5](https://github.com/wavyrai/tmux-ide) – tmux-ide adds an agent-aware terminal UI on top of existing tmux sessions, including live agent status, notifications, and crash-proof restore across clients.

* [agtop v0.6.1](https://github.com/ldegio/agtop) – Top-style terminal dashboard monitoring local AI coding agent sessions with live spend, token usage, context pressure, CPU, tool activity, and process details.

* [OtoDock v1.7.1](https://github.com/OtoDock/oto-dock) – Self-hosted multi-tenant platform for building and running collaborative company agents that work with Claude Code, Codex, Anthropic and OpenAI, and local models.

* [codemap v4.5.2](https://github.com/JordanCoin/codemap) – Generates compact, token-efficient architectural maps of codebases to give LLMs instant context.

* [codex-multi-auth v2.19.1](https://github.com/ndycode/codex-multi-auth) – Multi-account OAuth manager for the official Codex CLI with resilient routing, quota-aware rotation, and terminal-first diagnostics.

* [OwnMem v0.8.1](https://github.com/grpcer/ownmem) – Git-native memory that provides deterministic, reviewable local recall for AI coding agents, with evidence and risk gates to prevent harmful memory changes.

* [Metrik v0.21.0](https://github.com/keros68/metrik) – 桌面常驻工具用于统一查看本机各 AI 编程 Agent 的官方配额余量与 Token 消耗，支持 ChatGPT、Claude、GLM、Kimi 等。

* [CodeBurn desktop-v0.9.27](https://github.com/getagentseal/codeburn) – Interactive TUI dashboard that tracks token usage, cost, and performance across AI coding tools.

* [Superset desktop-canary](https://github.com/superset-sh/superset) – Terminal for running and managing 10+ parallel CLI coding agents locally.

* [Rome app-runtime-v0.7.2](https://github.com/rome-os/rome) – Agentic OS that provides a guardrailed environment for collaborative human and agent workflows with persistent tools, memory, and interfaces.

* [OpenAI API Server via Codex v0.2.2](https://github.com/hotchpotch/openai-api-server-via-codex) – Lightweight OpenAI-compatible API server that fronts Codex with optional API-key enforcement, enabling existing OpenAI clients to reuse code within Codex limits.

* [ThinkWatch Lite v2026.10.15](https://github.com/ThinkWatchProject/ThinkWatch-Lite) – Local gateway that connects AI clients once, records every request with cost and routing, redacts secrets, blocks dangerous tool calls, and enables safe upstream switching.

* [magpie v0.1.1165](https://github.com/yetone/magpie) – Desktop and terminal menu-bar app that lets you view and switch each installed AI agent’s configured model via a local model gateway.

* [Mixdog v1.0.12](https://github.com/tribgames/mixdog) – Windows desktop coding agent designed to match Codex performance while using substantially fewer tokens.

* [Mimi Remote v1.1.6](https://github.com/gaixianggeng/mimi-remote) – Native local-first iPhone and iPad workbench for managing Codex sessions on your Mac, including reviewing diffs, approving actions, and handling Git work.

* [AI Agent Session Center v2.10.54](https://github.com/coding-by-feng/ai-agent-session-center) – Local dashboard that monitors, controls, queues, and resumes Claude Code and Codex sessions with live terminals, tool logs, and 3D session robots.

* [SparkWork v0.14.14](https://github.com/alexanderizh/spark-agent) – Local-first desktop AI agent workbench for coding, research, documents, multimedia, and workflow automation with Claude/OpenAI/Anthropic support and SQLite memory.

* [agentbox v0.1.12](https://github.com/devilcoolyue/agentbox) – Self-hosted browser workspace for Claude Code and Codex CLI, offering persistent chat, terminal, files, Git review, and usage tracking via Docker sessions.

* [Wegent wework-v0.5.6](https://github.com/wecode-ai/Wegent) – Self-hostable platform for building and running AI agent teams across chat, coding, knowledge, and automation with shared capabilities and local execution via Wework.

* [Happy Coder cli-1.2.6-beta.1](https://github.com/slopus/happy) – Mobile and web client providing end-to-end encrypted remote access and control for Claude Code.

* [Vicoa v0.1.38](https://github.com/vicoa-ai/vicoa) – Vicoa is an AI orchestrator that runs and coordinates multiple coding agents across devices, providing a unified session dashboard with parallel worktrees and remote control.

* [Maestro-Flow v0.5.91](https://github.com/catlog22/maestro-flow) – Workflow orchestration framework that routes intent to adaptive multi-agent command chains, runs parallel agents, and provides a real-time dashboard and evolving knowledge graph.

* [Codex Mate v0.1.26](https://github.com/SakuraByteCore/codexmate) – Local-first AI control plane for Codex, Claude Code, and OpenClaw with one-click provider/model/profile switching, safe config writes, unified session browser, and MCP stdio.

* [VelaTerm v0.2.9](https://github.com/vlinx-io/VelaTerm) – A terminal manager that organizes AI coding sessions into a persistent project group session tree with agent-aware status and remote browser or SSH access.

* [LLMIO v0.9.4](https://github.com/atopos31/llmio) – Load-balancing gateway for multiple LLM provider APIs with REST compatibility.

* [Joanium v2026.1005.0](https://github.com/Joanium/Joanium) – Local-first AI desktop agent that runs scheduled background tasks, supports multi-model chat, and can use real integrations via MCP.

* [BaoCut agent skill baocut-v3.3.0-build....](https://github.com/JimLiu/baocut) – Agent Skill that lets coding agents drive the BaoCut local bcut CLI and Subtitle Studio to transcribe, translate, edit, and export subtitles from natural language.

* [Mesh Avatar Studio mesh-avatar@0.1.1](https://github.com/shinshin86/mesh-avatar-studio) – Mesh Avatar Studio turns a single illustration into an animated 2D mesh avatar with blinking, talking, head movement, and a local live-edit editor.

* [LiveAgent v2.0.0-beta.11](https://github.com/Stack-Cairn/LiveAgent) – Local-first AI agent desktop client combining multi-model chat with local tool execution, MCP tool bridging, and a remote gateway for access and collaboration.

* [Blave Agent desktop-v0.1.19](https://github.com/Blave-TW/blave-agent) – MacOS quant workspace that turns an agent prompt into versioned strategy code, runs validated backtests, and executes live trading without an LLM in the order loop.

* [Codex App Offline offline-v26.1007.231...](https://github.com/lusipad/unofficial-codex-app-offline) – Unofficial offline and portable repackaging for the OpenAI Codex Windows app with a local web gateway and Codex CLI.

* [AOCI-CODE v0.1.0-rc19](https://github.com/aoci-spec/aoci-code) – Indexing method that distills repository code and database knowledge into a portable, versioned cognition map for AI agents to read and act on consistently.

* [AxonHub v1.0.0-beta11](https://github.com/looplj/axonhub) – Modern AI gateway providing unified OpenAI, Anthropic, and AI SDK-compatible APIs with tracing, project management, and developer tools.

* [Mobile Dev for Codex v0.1.132](https://github.com/callstackincubator/codex-mobile-dev-plugin) – Codex plugin providing side-by-side live device streaming, control, logs, and performance monitoring for iOS simulators and Android devices.

* [AI Coding data-20261009-2240](https://github.com/panxunying/ai-coding-welfare) – Claude Code/Codex/Cursor welfare-style router list with free quota, model and pricing auto-updates plus one-click configuration scripts.

* [Hevy MCP Server hevy-mcp@6.1.21](https://github.com/chrisdoc/hevy-mcp) – Hevy MCP server that lets AI clients securely read, analyze, and write workout data, including routines and templates, via authenticated Hevy API requests.

* [AgentsDock v1.0.10-beta.4](https://github.com/ZhengyiLuo/AgentsDock) – An IDE client for agentic AI research that connects to a self-hosted backend to manage chats, run agent workflows, and review rich outputs.