Topics / #subagents

Subagents projects on GitHub

Collections of specialized AI agent subroles that work under a main assistant.

As of , the most-starred Subagents projects on GitHub are hermes-agent (254.8K stars), claude-howto (41.9K stars) and agents (41K stars). GitHub.Locker tracks 39 Subagents repositories with 598.4K stars combined. The fastest-growing on October 17 was hermes-agent, with +300 stars in a day.

39Projects
598.4KStars in total
+1.5KStars in a day
Browse in the catalog →github.com/topics/subagents ↗

Hot right now

Most stars gained in a day · Oct 17

+300

hermes-agent

NousResearch/hermes-agent · ★ 254.8K

Hermes Agent is a self-improving AI agent from Nous Research that creates and refines its own skills from experience, persists memory about you across sessions, and can be driven from a terminal or from chat apps like Telegram and Discord.

+207

vibe-coding-toolkit

soumatheusgomes/vibe-coding-toolkit · ★ 1.6K

A curated, battle-tested AI-coding workflow for Claude Code (and partly Codex): ready-to-copy plugins, subagent orchestration patterns, quality-gate configs, and prompt templates extracted from a real production fintech codebase.

+141

pstack-claude

michael-denyer/pstack-claude · ★ 2.8K

pstack is a port of Lauren Tan's (Poteto's) opinionated Cursor skill stack that improves coding-agent outcomes, adapted to run on Claude Code, Codex, Pi, Copilot, OpenCode, Gemini CLI, and Prime Agent.

+135

knowledge-work-plugins

anthropics/knowledge-work-plugins · ★ 30K

An open-source collection of 11 Anthropic plugins that turn Claude into a specialist for knowledge-work roles — sales, finance, legal, marketing, data and more — built for Claude Cowork and also compatible with Claude Code.

The all-time top

39 projects · by stars

  1. #1 · NousResearch/hermes-agent

    hermes-agent

    Hermes Agent is a self-improving AI agent from Nous Research that creates and refines its own skills from experience, persists memory about you across sessions, and can be driven from a terminal or from chat apps like Telegram and Discord.

    254.8K
  2. #2 · luongnv89/claude-howto

    claude-howto

    A free, open-source guide to Claude Code: 10 visual tutorial modules with Mermaid diagrams and copy-paste templates that take you from your first slash command to advanced agent setups in 11-13 hours.

    41.9K
  3. #3 · wshobson/agents

    agents

    A multi-harness marketplace of agentic building blocks for coding agents: 94 plugins, 202 agents, 183 skills, and 105 commands built from a single Markdown source and consumed natively by Claude Code, Codex CLI, Cursor, OpenCode, Antigravity CLI, and GitHub Copilot.

    41K
  1. 4

    oh-my-pi

    35.8Kcan1357/oh-my-pi · TypeScript

    omp (oh-my-pi) is a terminal coding agent — a fork of Mario Zetchner's Pi with the IDE wired in, built by Stencil Labs: 60+ model providers, 31 built-in tools, 14 LSP and 28 debugger ops, around 80k lines of Rust core.

  2. 5

    knowledge-work-plugins

    30Kanthropics/knowledge-work-plugins · Python

    An open-source collection of 11 Anthropic plugins that turn Claude into a specialist for knowledge-work roles — sales, finance, legal, marketing, data and more — built for Claude Cowork and also compatible with Claude Code.

  3. 6

    Claude-Code-Game-Studios

    26.8KDonchitos/Claude-Code-Game-Studios · Shell

    A template that turns a Claude Code session into a structured game studio: 49 specialized AI agents led by directors, 73 slash-command skills, 12 validation hooks, and 41 document templates.

  4. 7

    awesome-claude-code-subagents

    25.8KVoltAgent/awesome-claude-code-subagents · Shell

    A curated collection of 158+ specialized subagents for Claude Code: ready-made AI assistants, each tuned to one development task, organized into 10 categories from core development to data and AI.

  5. 8

    prime-agent

    22.1KPrimeIntellect-ai/prime-agent · Rust

    Prime Agent is an open-source coding and research agent for general and long-running work. It is built on two abstractions: the Recursive Language Model (RLM) programming model and a Continual Harness that keeps prompts, memories and skills as durable, refinable state.

  6. 9

    OpenHarness

    16KHKUDS/OpenHarness · Python

    OpenHarness is an open-source Python agent harness: the infrastructure that turns an LLM into a working agent with tool use, skills, memory and multi-agent coordination. It also ships ohmo, a personal agent that works for you in chat apps.

  7. 10

    Trellis

    15Kmindfold-ai/Trellis · TypeScript

    Trellis is an out-of-the-box engineering framework for AI coding. It persists specs, tasks, and project memory in your repo, so any coding agent works to your conventions instead of starting every session from scratch.

  8. 11

    gsd-core

    10.6Kopen-gsd/gsd-core · JavaScript

    GSD Core is a light-weight context-engineering and spec-driven development framework that drives AI coding agents like Claude Code, Codex, and Cursor through a disciplined phase loop.

  9. 12

    claurst

    10.3KKuberwastaken/claurst · Rust

    Claurst is an open-source, multi-provider terminal coding agent written in Rust. It began as a clean-room reimplementation of Claude Code and grew into a TUI pair programmer with a plugin system, chat forking, memory consolidation and a companion named Rustle.

  10. 13

    flue

    8.5Kwithastro/flue · TypeScript

    Flue is a TypeScript framework for building autonomous AI agents that get a secure sandbox, sessions, tools, skills, and filesystem access instead of just raw LLM API calls.

  11. 14

    superpowers-zh

    8.3KjnMetaCode/superpowers-zh · JavaScript

    The Chinese community edition of superpowers — a skills framework for AI coding tools. It installs 20 battle-tested workflow skills (brainstorming, planning, TDD, debugging, code review) into 26 AI coding agents so they follow proven work methods.

  12. 15

    awesome-codex-subagents

    6.4KVoltAgent/awesome-codex-subagents

    A curated collection of 170+ specialized subagent definitions for OpenAI Codex, covering development tasks like coding, infrastructure, security, data/AI, and developer experience.

  13. 16

    eve

    5.6Kvercel/eve · TypeScript

    eve is Vercel's open-source framework for building durable AI agents. It is filesystem-first: agent capabilities live in conventional file locations, so agent projects stay easy to inspect, extend, and operate.

  14. 17

    mistral-vibe

    5.1Kmistralai/mistral-vibe · Python

    Mistral Vibe is an open-source command-line coding assistant powered by Mistral's models that gives you a conversational, natural-language interface to explore, modify, and work with your codebase.

  15. 18

    KiroCrew

    4.5Kkirodotdev/KiroCrew · Python

    An open source, agent-powered development workspace that runs on your own hardware, locally or remotely. It persists between sessions, learns from corrections, and keeps working while you are away.

  16. 19

    claude-devtools

    4Kmatt1398/claude-devtools · TypeScript

    claude-devtools is a free, open-source debugging tool for Claude Code that reads the session logs already saved on your machine and reconstructs tool calls, thinking steps, subagent trees, token usage, and context window breakdowns in a visual desktop UI.

  17. 20

    qiushi-skill

    3.8KHughYau/qiushi-skill · JavaScript

    Qiushi-Skill is a collection of AI Agent Skills that distills one core principle (seeking truth from facts) and nine methodology tools from classical dialectical materialism and practical philosophy, aimed at making AI agents investigate first, focus on the main contradiction, validate through practice, and keep pushing until a task is actually finished.

  18. 21

    learn

    3.7Kamosblomqvist/learn · TypeScript

    'learn' is a personal AI learning system built by the author for the 'pi' agent tool - it packages a teaching philosophy, a visualization skill, quiz/question extensions, and research subagents into a '.pi' configuration that turns pi into a tutor.

  19. 22

    pstack-claude

    2.8Kmichael-denyer/pstack-claude · JavaScript

    pstack is a port of Lauren Tan's (Poteto's) opinionated Cursor skill stack that improves coding-agent outcomes, adapted to run on Claude Code, Codex, Pi, Copilot, OpenCode, Gemini CLI, and Prime Agent.

  20. 23

    PDoomVideo

    2.2KJohnHeibel/PDoomVideo · JavaScript

    The source code behind 'I'm Upping My P(doom)', a music video generated entirely by Claude Opus 5.5 in Claude Code — every scene, character and transition came from the model, not a human animator.

  21. 24

    dream-loop

    1.7Kachimala/dream-loop · JavaScript

    An agent skill that makes a coding AI build a game, app, or 3D scene with impressive visuals by looping: dream up a target screenshot with image generation, build toward it, have a critic subagent compare the live result to the target, and repeat until it matches.

  22. 25

    vibe-coding-toolkit

    1.6Ksoumatheusgomes/vibe-coding-toolkit · JavaScript

    A curated, battle-tested AI-coding workflow for Claude Code (and partly Codex): ready-to-copy plugins, subagent orchestration patterns, quality-gate configs, and prompt templates extracted from a real production fintech codebase.

  23. 26

    maki

    1.2Ktontinton/maki · Rust

    Maki is a terminal AI coding agent written in Rust, built to spend as few context tokens as possible, and extendable with neovim-style Lua plugins.

  24. 27

    autoprompt-skill

    1.2KSpielewoy/autoprompt-skill · JavaScript

    Autoprompt is a coding-agent workflow/skill that wraps your existing AI coding agent (Claude Code, Codex, etc.) and adds a review-fix-recheck loop, cutting task failures by about 45% in its own benchmark.

  25. 28

    second-brain-os

    1.1Kundefined-ui/second-brain-os · HTML

    A knowledge base an AI agent builds and maintains for you, in plain markdown files you own: what you read, watch and save becomes linked wiki pages you can query. The repo holds the full guide, a starter vault, agent skills and scripts.

  26. 29

    whip

    1.1Kcontext-labs/whip · Go

    whip is a fast coding-agent harness written in Go: an LLM runs a tool-use loop (bash, read, write, edit, subagent) inside an interactive bubbletea terminal session to work on your code.

  27. 30

    Step-Code

    950stepfun-ai/Step-Code · TypeScript

    Step Code is a terminal-based AI coding agent from StepFun: it handles the full task loop in your project — reading code, making changes, running tests — using Step models after you sign in.

  28. 31

    founder-skill

    923Jakeschincariol/founder-skill · Python

    Founder Skill is a set of eleven free Claude skills that let you stress-test a business idea before spending money on it: a board of directors modeled on Hormozi, Thiel and Jobs frameworks, a marketing director, a CFO, and a consumer panel of 100 simulated buyer agents.

  29. 32

    Orbital

    904zqiren/Orbital · Python

    Orbital is a local project agent: it keeps your project's state, decisions and history as plain files inside the project folder, so any coding agent — Claude Code, Codex, Cursor — picks up where the last one stopped without re-explaining anything.

  30. 33

    codeAgent

    884Harvil1/codeAgent · Python

    CodeAgent is an open-source AI coding assistant that lives in your terminal: you give it natural-language instructions and it reads your code, edits files, runs commands, searches for information, and finishes the task.

  31. 34

    Baize

    581Xu123-Bob/Baize · JavaScript

    Baize is an open-source AI coding agent CLI that acts as a drop-in alternative to Claude Code CLI, letting you describe tasks in natural language while it writes, edits, and runs code for you.

  32. 35

    open-pstack

    426ericlitman/open-pstack · TypeScript

    Open Pstack ports Lauren Tan's (@poteto) pstack — engineering rules, workflows, skills, and small local tools for coding agents — from Cursor to Claude Code and Codex.

  33. 36

    pi-optchat

    267jonaslsaa/pi-optchat · TypeScript

    pi-optchat is an extension for the Pi coding agent that gives it persistent, endless memory per profile using a summary tree instead of context compaction, plus background subagents and the ability to import history from other agents.

  34. 37

    astra-advisor

    242DannyMac180/astra-advisor · Python

    Astra Advisor is a Codex plugin for capability-routed software delivery: GPT-6 Astra plans the work, delegates bounded deliverables to native subagents when useful, and owns verification and acceptance.

  35. 38

    oar

    198botiverse/oar · TypeScript

    OAR (Open Agent Runtime) is a provider-independent programming interface for coding-agent harnesses like Claude Code, Codex and Cursor, so you build one integration instead of writing separate harness logic for each agent.

  36. 39

    oh-my-subagents

    191ringlochid/oh-my-subagents · Python

    A local orchestration runtime for Codex and Claude subagents: durable AI teams with persistent task state, reusable responsibility trees and interruption recovery, instead of ad-hoc delegation.

Often tagged together