Coding agents & developer tools
Autonomous coding agents
4,327 agents · most starred first
Autonomous coding agents, page 16 of 44
Autonomous multi-agent coding system on local LLMs. No frameworks, no API costs. Plan → Build → Test → Fix with Qwen 80B + 7B on your own GPUs.
Superzent - an agent-first workspace built on Zed
Lightcode is a coding agent written in Go.
Ultracode making for OpenCode, inspired by Claude Code. Model fusion for open-weight models to beat frontier models at scale, cost and quality.
The loop never sleeps — you finally can. Agents automate the entire development lifecycle (write, review, merge, learn) end to end, with a published safety record. Self-host today; rentable hosted loops in development. Any GitHub repo; Gittensor-native intelligence built in.
No API keys required. Transform the native ChatGPT, Claude, and Gemini web interfaces into fully autonomous local coding agents.
Agentic coding in the terminal: the model writes its own multi-agent workflows (Gigacode). Provider-agnostic, local-first, 15+ model providers, MCP support.
User Interface for Prime-Agent
VinusCLI - the AI coding agent in your terminal. Docs, issues and releases. Ships via npm; source is closed.
AI-powered CLI for automated code and framework migrations using autonomous multi-agent orchestration.
🤖 World's first AST-powered AI development agent | Autonomous Django/Flask builder | Zero-token intelligence | Military-grade sandbox | Made in India 🇮🇳
GPT-4 (prompt) AI assists in writing, fixing, or reviewing code by providing suggestions, guidance, and explanations. It can also serve as a pair programming buddy to enhance your coding experience and help create or explain tasks.
Pythoness: use natural language to define Python functions.
📕Copy-Paste PROMPTS for deploying 🤖Autonomous AI Agents for Codebase Optimization➡️ Deploy specialized AI agents that work 24/7 to optimize, secure, and enhance your codebase
AI-powered development orchestrator: Supports Claude Code, Codex and Langchain Deepagents + Openrouter backend
An intelligent orchestration system that automates the software development lifecycle from Feature ideation through code delivery using AI-powered planning and execution.
Air-gapped coding agent in one Rust binary: a model on your own GPU (Ollama/LM Studio), --offline enforced by tests. Ships Q56, a hidden-test benchmark for local models.
Let Codex test multiple fixes in isolated worktrees, compare evidence, and apply only the safest proven solution.让 Codex 不再盲改代码,而是并行测试多个方案,用证据选出最稳的修复。
Parallel coding agent workspace with native terminals and git worktrees.
An autonomous AI coding agent with novel innovations in tool state management and code editing, running in a Docker sandbox with a persistent shell, parallel tools, and a VNC desktop.
A lightweight (10 MB) AI IDE alternative to Cursor and Windsurf
transformers gguf llama text-generation unsloth base_model:ByteDance-Seed/Seed-Coder-8B-Reasoning
An AI coding agent for enterprise development, combining project-aware reasoning with human collaboration.
Your AI development team — 27+ agents and 21 skills that build, review, debug, and ship production code. Works with Claude Code, Cursor, VS Code, Windsurf, Codex, and OpenCode. 7 framework packs: Vue, React, Next.js, SvelteKit, Angular, Astro, Nuxt.
Local-first desktop command center for Claude Code, Codex, OpenCode, and other coding agents—parallel tasks, worktrees, dev services, port forwards, diffs, and review.
Think first, then code. Where Claude Code and Codex jump straight to writing, this stays review-first: a code reviewer, security scanner, root-cause debugger, and code generator, all in one shell-native loop.
Open-source repository standard, starter template, validator, and multi-agent workflow for making vibe-coded and AI-generated software understandable, testable, reviewable, and maintainable. Codex-first reference implementation.
MusePi — desktop-first AI coding assistant: Electron GUI + TUI/CLI + daemon + desktop companion. Native video understanding, MCP, advisor, memory, multi-domain skills; own update channel.
an open experimental coding harness
Open coding agent that matches the best at lower cost: any model, sandboxed, scriptable, embeddable. Engine of Murage.
Caravel — a local-first workspace for AI coding agents. Delegate tasks, follow progress, and review code across projects.
A native desktop coding agent built with Rust and GPUI.
Open-source Windows AI IDE built with .NET 10 WPF, featuring a fine-tuned local AI router (1.5B), live Blender & Unity copilots, self-healing agent loops, and multi-tool execution.
A senior-engineer protocol for polyglot code generation, architecture, testing, security, performance, and agent validation.
Agentic Spec-Driven Development — 13 agents, 58 MCP tools, 10-phase enforced pipeline. From meeting transcript to PR in one command. Install via npm install -g specky-sdd@latest
OPENCODE APK ANDROID 📱 Native AI coding agent + On-device APP FACTORY + PERSISTENT MEMORY 🚀 PLUG & PLAY · FULL DEVICE CONTROL · NO ROOT REQUIRED.
Free AI coding assistant for your terminal and browser. Runs on free models or your ChatGPT, Claude, Grok, Gemini or Copilot plan. No account needed.
Open-source desktop coding agent for macOS. Use any model: local with Ollama, your own API keys or a ChatGPT sign-in. Parallel chats, MCP connectors, skills, and computer use on a desktop of its own.
GatherThread 共序:本地 AI 编程 Agent 多人联机协作、实时共享上下文。Self-hostable collaboration for local AI coding agents, Codex Desktop connector & DeepSeek Harness plugin.
generate open-source project ideas for developers, when you're bored and want to work on something
Give AI coding agents (Claude Code, Cursor, Aider, Codex) a structured autonomous loop with guardrails — boundaries, 5 verification gates, 3-layer self-reflection, and autonomous remediation. pip install ouro-loop. Zero dependencies.
SAM - Autonomous TDD agent system for Claude Code, Cursor, Gemini CLI, GitHub Copilot, and Antigravity. PRD in, working tested app out.
YzPzCode - The Unified Workspace for AI Coding Agents
A self-hosted software factory — define goals, agents decompose and execute them through enforced pipelines, with humans in the loop where it matters.
You pick the model per task, with review, evidence, and a PR at the end.
AI-native methodology for Go backend microservices: rebuild the whole codebase from docs/ + .claude/skills, driven by a 12-lens high-performance architecture decision framework with mechanical doc↔code sync.
Provider-agnostic AI dev pipeline: clarify → plan → build → review → PR across your repos, mixing LLM providers per stage with your own keys. No vendor lock-in, no markup.
Local-first terminal agent runtime with isolated subtasks and intent-aware safety.
One Works application monorepo for AI agents, plugins, desktop, web, CLI, and Relay.
Ultra cost-effective terminal AI coding assistant with excellent token cache hit rate. Built in ~10K lines of Go, it uses DeepSeek by default with multiple modes and sandbox.
Open-source local-first dispatcher for AI coding agents. Verification-gated, BYOK, rolls back on test failures.
Vis is a coding agent that combines tools into Python programs. It can chain operations, run them in parallel, and inspect results before deciding what belongs in the conversation.
Think first, then code. Where Claude Code and Codex jump straight to writing, this stays review-first: a code reviewer, security scanner, root-cause debugger, and code generator, all in one shell-native loop.
A native macOS code editor for CLI-agent workflows. Split editor and terminal panes, a cross-project Agent Inbox, private notifications, LSP, Git context — zero Electron.
The agent-first IDE, built for production.
A lightweight coding agent for teaching, research, and experimentation
The operating system for autonomous software engineering.
Workspace-first AI coding hub for managing agents, reviews, terminals, and task flows across multiple providers.
The factory that builds software — and itself. A runtime for self-improving agentic loops: the tracker is the control plane, git is the truth, CI is the gate.
Native Windows GUI for Oh My Pi: sessions, models, providers, MCP, plugins, context, and browser tools.
The pit crew for agentic coding.
Free desktop app to run AI coding agents in parallel: Claude Code, Codex, Cursor, OpenCode, Kimi, Grok. Worktree per thread, shared memory.
Self-hosted workspace that runs several AI coding CLIs (OpenCode, Codex, Claude Code, Gemini) side by side in persistent rooms.
OSC Agent - An open-source contributor CLI agent
Educational Python coding agent with provider-agnostic LLM support, skills system, and extensible tool suite.
A multi-provider AI coding agent with the persona of a Tech-Priest
Agentic coding harness with persistent memory and a REPL body. Built on Ori Mnemos. Open source must win.
Asynchronous Software Development
Autonomous multi-agent orchestrator for AI coding agents via tmux
Agentic Spec-Driven Development — 13 agents, 58 MCP tools, 10-phase enforced pipeline. From meeting transcript to PR in one command. Install via npm install -g specky-sdd@latest
Lionclaw is a local control plane for AI coding agents. It runs AI agents as durable, auditable workers with explicit state, skills, channels, schedules, and bounded permissions.
ClosedLoop.AI — The open-source workspace for team-based agentic software development.
An autonomous AI agent plugin for Acode that enables LLMs to directly read, create, and refactor files, while executing terminal commands through a secure agentic loop—bringing desktop-class AI automation to Android.
Run and operate Claude Code on your own machine — fully remotely, from your phone or any browser. Self-hosted server + installable PWA that drives the real claude CLI with your Claude subscription.
Private, offline AI coding agent for local Qwen models via Ollama — zero dependencies, real tools, multi-agent. Your code never leaves your machine.
transformers safetensors llama text-generation conversational base_model:deepseek-ai/deepseek-coder-6.7b-instruct
transformers pytorch mistral text-generation code en
gguf ravenx openfable soul-infusion qwen3_6 coding
Model Optimizer safetensors kimi_k25 nvidia ModelOpt Kimi-K2.7-Code
Self-hosted mission control for real Claude Code and Codex sessions — persistent terminals, automations, and browser/mobile control.
A coding assistant built on persistent multi-agent teams: durable members that discuss, chat, and re-sync continuously — Discuss -> Code in parallel -> Sync -> Verify together.
Kimi Code-style agent orchestration harness for the Pi coding agent — Swarm, Goal, Plan, Permission, Task, Hooks, Skills, TUI
Open-source no-progress governor for AI coding agents. Detect repeated work that changes nothing. Shadow first, enforcement only after evidence.
A systems language designed so an AI agent can safely write and run backend code: proven no GC/races/overflow, GBNF-constrained generation, structured diagnostics, OS-process sandboxing.
Autonomous AI coding agent orchestrator. Watches GitHub issues, spawns Claude Code, Codex CLI, or Antigravity CLI in isolated git worktrees, opens PRs, and runs an independent reviewer pass. Self-hosted, Python, multi-repo.
Lab Agent — open-source coding agent (terminal + desktop)
A coding workspace for AI agents, with a desktop app, terminal CLI, and shared tasks.
A lightweight, inspectable TypeScript coding agent for local software-engineering work.
Another agent coding tool with auto mode, auto routing and more
Local desktop coding assistant you can run with any local model. No API key required. It can read your project, edit files, run commands, verify results, and show you exactly what changed and scrape the internet. Private remote access incl. sharing files across devices e.g. sending files to your mobile phone. (Min Requirements 12GB VRAM / 48GB RAM)
Desktop coding agent app for macOS and Linux. Any model, on your code and UI. Built on DeepSeek Harness.
Oh My Diff: A Moddable Codex Runtime
Unofficial: Meta's Muse Spark as one coding agent in every editor: Zed, JetBrains IDEs, Neovim and Emacs over ACP, plus Cursor, Windsurf, VSCodium and VS Code. Subagents on a live map, browser checks on your dev server, ESP32 hardware skill, MCP, voice, 15 languages. Muse Code CLI or Model API. MIT.
The coding agent that keeps working. Terminal, web, phone, and VS Code.
Run a team of Claude Code agents in one native window (Rust, no Electron): a shared task board and memory over MCP, an orchestrator that runs workers in git worktrees and lands what they deliver, and Hangar Mobile to steer them from your phone. Claude Code only. Local, open source.
A genie you ask to make your wishes come true, with the help of Claude Code, of course
Autonomous ChatGPT AI experiment written in Delphi
Control OpenAI Codex coding agents from your phone. Run multiple agents, get push notifications, and reply from your lock screen.
DeepNext is an AI-powered junior software engineering assistant designed to automate the process of converting issue descriptions into actionable pull requests.