Upstar existe aussi en français.

Voir cette page en français

Upstar ranks the fastest-rising GitHub projects every hour, across 69 topics. Browse topics

#coding-agents · AI & ML

Coding agents

Codex, Cursor, Gemini CLI, Aider and other agents that write code for you, with the plugins and instruction files around them.

This week (Nov 1 – Nov 7), the top-ranked are dsh-superagent-prototype (+18,033), skills (+2,188) and dsh-Go (+10,509). The 437 ranked repos gained 144,689 stars, 18% more than the week before.

Updated Nov 7, 16:39 UTC · ranked every hour from GitHub’s daily star counts

Updated 8 min ago

GitHub topic ↗ (opens in new tab)

Emerging this week

Showing 50 of 58

How Upstar ranks

Listed by acceleration against their usual week; the score is the rankings’ own: stars gained in 7 days ÷ √(stars before, minimum 100).

  1. 1
    feimacode/open-design-agent-kit Make posters, social posts, slides and landing pages in the coding agent or chat you already use (Copilot, Claude, Codex, Cursor), with 198 design systems, checks and PDF/PNG/PowerPoint export. Breakout week (~1 a week) HTML HTML · 1k total Spotted at 478 stars
    Gained in 7 d +400 Score 15.8 Total 1k
  2. 2
    Muvon/octomind Open-source AI coding agent and agent runtime: one binary, any model, MCP-native. Runs in terminal, CI, or as a daemon. 50× its usual pace (~8 a week) Rust Rust · 596 total Healthy
    Gained in 7 d +397 Score 28.1 Total 596
  3. 3
    YizhengWw/HachimoDock A desktop pet that reacts to your Claude Code sessions in real-time — thinking, typing, juggling, sleeping, and more. 76× its usual pace (~5 a week) C C · 933 total Healthy Spotted at 298 stars
    Gained in 7 d +379 Score 16.1 Total 933
  4. 4
    guregodevo/memdoor AI coding agent that makes a team of small models beat a frontier one Breakout week (~1 a week) Go Go · 374 total
    Gained in 7 d +332 Score 33.2 Total 374
  5. 5
    dstrml/wu The fast, native code editor and agent workspace. 36× its usual pace (~9 a week) Rust Rust · 535 total Healthy
    Gained in 7 d +323 Score 22.2 Total 535
  6. 6
    solongate/agent-security Local-first, open-source security gateway for AI coding agents. Enforce tool-call policies, detect secrets, and audit actions on-device. Breakout week (~2 a week) Go Go · 428 total
    Gained in 7 d +289 Score 24.5 Total 428
  7. 7
    MicTx/spec-harness 面向 AI 编程智能体的任务包工作流工具，用可验证的范围、证据和 Git 门禁管理长期改动。(Task-package workflow for AI coding agents with verifiable scope, evidence, and Git gates.) 91× its usual pace (~3 a week) Python Python · 283 total
    Gained in 7 d +272 Score 27.2 Total 283
  8. 8
    uniquechao/PandaDOS Your self-hosted control plane for Claude Code and Codex. Turn issues into queued, observable, approval-aware coding workflows—from any browser. After quiet weeks TypeScript TypeScript · 363 total Mixed signals
    Gained in 7 d +193 Score 14.8 Total 363
  9. 9
    FinanceFlash/unvibecode Complex code hides connections, workflows, and business risks.Unvibe complex code. Trace business workflows. 19× its usual pace (~24 a week) Jupyter Notebook Jupyter Notebook · 1.3k total Healthy Spotted at 448 stars
    Gained in 7 d +463 Score 16.1 Total 1.3k
  10. 10
    wanshuiyin/HERO-Anti-OverDefense HERO = Hashing · Edge cases · Rubrics · Overbuild — the four shapes coding agents over-defend in. A paste-in contract that stops them. Works with Claude Code, Codex, Antigravity, Cursor, Copilot, Windsurf, Gemini CLI. 19× its usual pace (~14 a week) Python Python · 800 total Healthy
    Gained in 7 d +264 Score 11.4 Total 800
  11. 11
    KenKaiii/gg-framework The coding agent that gets more out of your AI subscription. Fastest and leanest of 20 agents tested, with unlimited local search of real open-source code. After quiet weeks TypeScript TypeScript · 235 total Healthy
    Gained in 7 d +180 Score 18.0 Total 235
  12. 12
    YCC-Lover/research-starter-skills 面向科研新手的 Codex 科研协作技能集：选题、文献检索与阅读、论文写作、自然润色、审稿返修和可编辑学术汇报。结合 Research-Starter-Kit 原创教程，用 1 个总入口与 6 个专业模块推进研究。 Research training, writing & editable presentations for Codex. 33× its usual pace (~4 a week) Python Python · 185 total
    Gained in 7 d +132 Score 13.2 Total 185
  13. 13
    runloom/pier 几个 AI 终端开着，谁在等你一眼能看见。Claude Code、Codex 还在原来的终端里，旁边就能改文件、看 Git。macOS 本地工作台。 17× its usual pace (~7 a week) TypeScript TypeScript · 561 total Healthy
    Gained in 7 d +121 Score 5.8 Total 561
  14. 14
    skaft-software/octet fast rust coding agent, extensible in any language, multimodal inputs, deep provider integrations. After quiet weeks Rust Rust · 114 total Mixed signals
    Gained in 7 d +104 Score 10.4 Total 114
  15. 15
    niuwoai/willdeep-cli 自托管 AI Coding Agent（Rust）：常驻 Runtime 撑起终端 / 浏览器 / 手机三端；每行改动可追溯到具体 Agent 与工具调用，子 Agent 的活由退出码裁决，检查点回退、目标 token 预算与验收清单跨轮次跨重启累计，用量与反馈账本留在本地。接任意 OpenAI 兼容或 Anthropic 端点。Self-hosted auditable AI coding agent: per-line attribution, exit-code-verified sub-agents, restart-safe goal budgets, local usage… After quiet weeks Rust Rust · 107 total Healthy
    Gained in 7 d +94 Score 9.4 Total 107
  16. 16
    SLEEPYBQ/MemoSync Co-manage memory with your coding agent across sessions. A local workbench for Claude Code, Codex and Pi. After quiet weeks TypeScript TypeScript · 186 total Mixed signals
    Gained in 7 d +87 Score 8.7 Total 186
  17. 17
    jiucai0915-sys/deepcode DeepSeek-native coding agent for your terminal | DeepSeek 原生终端编程 Agent After quiet weeks TypeScript TypeScript · 231 total
    Gained in 7 d +81 Score 6.6 Total 231
  18. 18
    Endokelp/Endoplexity Agentic browser control driven by your own Claude / Cursor subscriptions After quiet weeks TypeScript TypeScript · 413 total Low activity
    Gained in 7 d +81 Score 4.4 Total 413
  19. 19
    taskless/cli Taskless CLI and Agent Skills After quiet weeks TypeScript TypeScript · 141 total Healthy
    Gained in 7 d +75 Score 7.5 Total 141
  20. 20
    ThiloReintjes/tinybar A lighter, from-scratch alternative to CodexBar: Claude, Codex, Gemini, Cursor, OpenCode Go and Copilot limits, token history and API-equivalent cost. No WebKit, read-only logins. 74× its usual pace (~1 a week) Swift Swift · 80 total
    Gained in 7 d +74 Score 7.4 Total 80
  21. 21
    Tencent/LoopForge Self-hosted workflows for coding-agent delivery. Orchestrate CodeBuddy, Codex and Claude Code with scripts and human approval gates. Includes Local workflows for coding assistants. 14× its usual pace (~5 a week) Go Go · 144 total Healthy
    Gained in 7 d +71 Score 7.1 Total 144
  22. 22
    mauhpr/agentlint Local guardrails for AI coding agents. Check proposed file changes and shell commands, block risky patterns, and return actionable feedback. Works with Claude Code, Codex, Cursor, Gemini CLI, and more 70× its usual pace (~1 a week) Python Python · 133 total Mixed signals
    Gained in 7 d +70 Score 7.0 Total 133
  23. 23
    dengmengmian/CodeLeveler A local-first coding agent that works in your repository and leaves reviewable changes After quiet weeks Rust Rust · 162 total
    Gained in 7 d +64 Score 6.4 Total 162
  24. 24
    shenysun/skills-manager CLI-first hub for agent skills: one npx install, manage in chat, distribute to 70+ agents 62× its usual pace (~1 a week) TypeScript TypeScript · 242 total Mixed signals
    Gained in 7 d +62 Score 4.6 Total 242
  25. 25
    lobehub/acceptance ✅ Acceptance is open-source e2e testing for Agents. Your coding agent turns a prompt into observable checks, runs the real product, and publishes a report with screenshots, recordings, and logs attached to each check. You can accept the work, or send individual checks back. 18× its usual pace (~3 a week) JavaScript JavaScript · 90 total
    Gained in 7 d +53 Score 5.3 Total 90
  26. 26
    callstackincubator/codex-mobile-dev-plugin A Codex plugin for a better mobile development experience for android and iOS 6× its usual pace (~9 a week) C++ C++ · 224 total
    Gained in 7 d +53 Score 4.1 Total 224
  27. 27
    Eslamasabry/opencode-mobile-next Run Claude Code and OpenCode on your Android phone, no computer needed. Or follow and steer agents on your computer: live sessions, tool approvals and diffs. 25× its usual pace (~2 a week) Dart Dart · 67 total Healthy
    Gained in 7 d +50 Score 5.0 Total 67
  28. 28
    jeffreycayao/Dev-Labs DevLabs™ Free, open source and local-first. A multi-agent harness that works with the AI subscriptions you already pay for. It turns the terminal coding CLIs you already run into a team:. After quiet weeks TypeScript TypeScript · 57 total
    Gained in 7 d +50 Score 5.0 Total 57
  29. 29
    galiprandi/job-seeker Automate your job search with your favorite coding agent. Skills, guides, and scripts that any agent (Devin, Claude, Cursor, opencode) consumes to search, apply, and track jobs on your behalf. 16× its usual pace (~3 a week) JavaScript JavaScript · 112 total Healthy
    Gained in 7 d +49 Score 4.9 Total 112
  30. 30
    cluster1900/apexo-code Apexo: An open-source, general-purpose coding agent for terminal (TUI), web, and desktop with autonomous goals, workflows, persistent memory, command sandboxing, and first-class support for Gemini, Grok, and OpenAI-compatible models. After quiet weeks TypeScript TypeScript · 303 total Mixed signals
    Gained in 7 d +47 Score 2.9 Total 303
  31. 31
    jossuealcacao-exe/madre MADRE · Codex, Claude Code, Gemini CLI and OpenCode in one local chat on your machine, with one memory every agent shares. Home for all your projects and a Coordinator above them. Read-only by default; you unlock more per message. In Spanish and English. Public beta. 5× its usual pace (~10 a week) JavaScript JavaScript · 120 total
    Gained in 7 d +46 Score 4.6 Total 120
  32. 32
    tommy0103/obelisk Every past session, subagent, and workflow -- queryable by your agent, browsable by you 4× its usual pace (~31 a week) JavaScript JavaScript · 906 total Healthy
    Gained in 7 d +126 Score 4.5 Total 906
  33. 33
    delta-box/deltabox-lite Lightweight, agent-native sandboxing with diff, commit and abort After quiet weeks Rust Rust · 321 total Healthy
    Gained in 7 d +37 Score 2.2 Total 321
  34. 34
    ericrisco/rsc-harness Your agent invents things because it has no memory, and can't touch your database because it has no arms. rsc is the meta-harness that gives it both, plus the trade to know the job and the reflexes to not break your repo. For Claude Code, Codex and 15 more. 4× its usual pace (~34 a week) JavaScript JavaScript · 532 total Healthy
    Gained in 7 d +122 Score 6.0 Total 532
  35. 35
    Get-Concord-AI/concord-mcp Live messaging for coding agents 3× its usual pace (~57 a week) TypeScript TypeScript · 993 total Healthy
    Gained in 7 d +194 Score 6.9 Total 993
  36. 36
    rsaryev/localix Open source Agentic Development Environment 4× its usual pace (~8 a week) TypeScript TypeScript · 57 total
    Gained in 7 d +33 Score 3.3 Total 57
  37. 37
    Gang-of-Beads/pi-web Web UI for Pi Coding Agent that keeps sessions alive in real workspaces. 11× its usual pace (~3 a week) TypeScript TypeScript · 115 total Healthy
    Gained in 7 d +32 Score 3.2 Total 115
  38. 38
    AxiomCodeAI/axiomcodegraph Turn any codebase into a knowledge graph grounded in formal methods. Your AI agents get exact answers for any code-related task, ensure correctness, and optimize time and token usage. 3× its usual pace (~39 a week) TypeScript TypeScript · 516 total Healthy
    Gained in 7 d +122 Score 6.1 Total 516
  39. 39
    lllleolin-max/VibeGit Local-first desktop safety net for AI-assisted coding. Create checkpoints, inspect changes, safely restore versions, and back up projects to GitHub—without destructive Git commands. 31× its usual pace (~1 a week) TypeScript TypeScript · 297 total Mixed signals
    Gained in 7 d +31 Score 1.9 Total 297
  40. 40
    AndrewDryga/coop Leverage full potential of coding agents without the risks. Run them in a sandbox with restricted networking, file access and secrets hidden. By Protectorate. After quiet weeks Go Go · 60 total Healthy
    Gained in 7 d +30 Score 3.0 Total 60
  41. 41
    shx2005/codex-account-switcher 你是否厌倦了 Codex 多个账号之间来回切换？本地账号管理、官方登录、额度查询、快速切换与跨设备导入导出。Local Codex account profiles, login, usage, switching, and export/import. 3× its usual pace (~95 a week) Python Python · 937 total Spotted at 190 stars
    Gained in 7 d +282 Score 11.0 Total 937
  42. 42
    cocacola-lab/awesome-embodied-rsi A curated list of Embodied Recursive Self-Improvement (Embodied RSI) research, systems, benchmarks, and industry updates. 3× its usual pace (~12 a week) 275 total Mixed signals
    Gained in 7 d +35 Score 2.3 Total 275
  43. 43
    seungdori/web-stylebook-mcp MCP server that gives AI coding agents scored design directions, UI-state plans, and WCAG-checked design tokens. Deterministic, offline, no API key. After quiet weeks TypeScript TypeScript · 32 total Healthy
    Gained in 7 d +28 Score 2.8 Total 32
  44. 44
    Sanuu7/AndroidHarness Android Harness AI Coding Agent Native app for mobile, No Root No PC 3× its usual pace (~11 a week) JavaScript JavaScript · 160 total Healthy
    Gained in 7 d +30 Score 2.6 Total 160
  45. 45
    prasanthsasikumar/openclicky Free, open-source voice assistant for macOS: ask about your screen and a pointer shows you where, dictate anywhere (offline by default), hand tasks to a Codex agent. Your own key or a free account. MIT. 14× its usual pace (~2 a week) C# C# · 47 total Mixed signals
    Gained in 7 d +27 Score 2.7 Total 47
  46. 46
    ozkancirak/Ration AI coding quotas in your Windows tray. 14× its usual pace (~2 a week) C# C# · 71 total
    Gained in 7 d +27 Score 2.7 Total 71
  47. 47
    imMamdouhaboammar/chatgpt-codex-plugin-autopilot Self-hosting ChatGPT/Codex Plugin autopilot — copy-in, repo-owned. Converts agentic repos into validated, deterministically packaged Plugins. 3× its usual pace (~12 a week) Python Python · 123 total Mixed signals
    Gained in 7 d +30 Score 3.0 Total 123
  48. 48
    AlysisAi/AlysisCode Open-source, multi-model agent for coding, research, and everyday workflows. Independently built engine with our own Alysis Sandbox, strict command isolation by default, and local model support. 4× its usual pace (~6 a week) Python Python · 212 total Healthy
    Gained in 7 d +24 Score 1.8 Total 212
  49. 49
    Priyanshu-byte-coder/contextrot Find out where your coding agent starts degrading. Personal context-rot analytics from your own sessions - 100% local, zero-config: uvx contextrot 5× its usual pace (~5 a week) Python Python · 114 total Mixed signals
    Gained in 7 d +23 Score 2.3 Total 114
  50. 50
    termio-sh/termio A terminal-first agentic development environment for agentic coding. Build for CLI/TUI agent. Runtime for Coding Agent, Tmux alternative 2× its usual pace (~22 a week) Swift Swift · 785 total Healthy
    Gained in 7 d +50 Score 1.8 Total 785
Show more

Back to the ranking Past weeks: Oct 26Oct 19

About this topic

Some of these agents work inside an editor, others in a terminal, OpenCode, Cline, Roo Code and GitHub Copilot among them; the projects built around them also include harnesses that drive an agent. Claude Code and its add-ons have their own topic. AI agents covers agents for any task, while this one keeps to those that program. For developers comparing agents, or after instructions and add-ons for the one they already use.