Ecosystem
Our picks from the AI coding landscape. Every tool here is one we have used, evaluated, or closely follow. Opinionated, not exhaustive.
AI Coding Agents
11Terminal-native coding agents and assistants
Claude Code
Anthropic's agentic coding tool. Lives in your terminal, understands your codebase, executes tasks through natural language. The one we built ClaudePedia to document.
GSD 2
A heavily modified Pi Code fork with GSD embedded. Meta-prompting, context engineering, and spec-driven development: agents work autonomously without losing the big picture.
OpenCode
Provider-agnostic terminal coding agent with TUI and LSP-aware editing. Now the dominant 'OpenCode' fork with 156K stars; the original Go fork was eclipsed.
Kilo Code
All-in-one agentic engineering platform. #1 coding agent on OpenRouter with 1.5M+ users and 25T+ tokens processed.
Gemini CLI
Google's open-source AI agent for the terminal. Brings Gemini directly to your command line with MCP support and 1M token context.
Codex CLI
OpenAI's lightweight coding agent for the terminal. Written in Rust for speed.
Aider
Terminal-native pair programmer that edits code in your local git repo. Maps repo structure, applies diffs, auto-commits each change. Pioneered the 'AI in your git repo' pattern.
Goose
Block's open-source on-machine AI agent. Extensible via MCP, runs locally, supports any LLM provider. The only major Rust-based agent in active production use.
Crush
Charm's terminal coding agent with polished TUI, multi-model support, and LSP integration. Successor to the original OpenCode (Charm).
OMO (oh-my-openagent)
The leading OpenCode plugin (formerly oh-my-opencode). Adds Sisyphus orchestration, async subagents, curated tools, and a Claude Code compatibility layer to OpenCode.
SuperClaude Framework
Configuration framework that adds 25+ slash commands, personas, and MCP integrations on top of Claude Code. The largest community framework layered on Claude Code.
AI-Native IDEs
6Integrated development environments with built-in AI agents
Antigravity
Google's agentic development platform. Agent-first IDE with a Manager Surface for spawning and orchestrating multiple agents across workspaces. Free in public preview.
Windsurf
AI-powered IDE by Codeium. Deep codebase understanding with Cascade, their agentic flow engine.
Cursor
The AI code editor that started the agentic IDE category. Composer mode, multi-file editing, strong model support.
Zed
High-performance, multiplayer code editor in Rust from the creators of Atom and Tree-sitter. First-class agent panel and MCP support. The mainstream OSS AI-native editor.
Cline
Largest agentic VS Code extension. Plan/Act modes, MCP support, file edits and terminal commands with human-in-the-loop approval.
void
Open-source VS Code fork positioned as a transparent alternative to Cursor. Native agent and model-routing built in.
Personal AI Assistants
8General-purpose conversational agents and chat platforms with persistent memory
OpenClaw
Personal AI assistant that runs on your own devices and answers via the channels you already use (SMS, Telegram, Discord, etc.). 369K stars in 6 months — the breakout personal-AI repo of 2026.
Hermes Agent
Nous Research's self-improving agent. Persistent memory, builds skills from experience, lives on 15+ platforms (CLI, Telegram, Discord, Slack, WhatsApp, Signal, email, SMS) from one gateway.
Open WebUI
User-friendly AI interface supporting Ollama, OpenAI API, and dozens of others. The dominant open-source ChatGPT-style UI.
LobeHub
Agent teammate platform for work and life. Multi-agent collaboration, agent team design, agents as the unit of work interaction.
AnythingLLM
All-in-one AI productivity accelerator. On-device and privacy-first with no annoying setup. Document chat, agents, and tools in one app.
LibreChat
Enhanced ChatGPT clone with first-class MCP support. Agents, model switching, code interpreter, langchain, OpenAPI Actions, secure multi-user auth, self-hostable.
Jan
Open-source alternative to ChatGPT that runs 100% offline on your computer. Strong local-first focus.
Khoj
Self-hostable AI second brain. Get answers from the web or your docs, build custom agents, schedule automations, do deep research. Turns any LLM into your personal autonomous AI.
Agent Frameworks
15Libraries and platforms for building AI agents and pipelines
DeepAgents
Agent harness built with LangChain and LangGraph. Planning tool, filesystem backend, subagent spawning: handles complex agentic tasks out of the box.
LangGraph
Build resilient language agents as graphs. Stateful, multi-actor workflows with cycles, persistence, and human-in-the-loop.
LangChain
The agent engineering platform. Composable chains, tool use, retrieval, memory: the foundation everything else builds on.
CrewAI
Role-based multi-agent orchestration. Agents take on personas (researcher, writer, etc.) and collaborate with minimal boilerplate. Native MCP and A2A support.
OpenAI Agents SDK
Official OpenAI framework for multi-agent workflows. Lightweight, built-in MCP tool integration, persistent sessions, and realtime voice agents.
Claude Agent SDK
Wraps the Claude Code CLI as a programmable SDK. Bidirectional interactive conversations, custom tools and hooks as functions. Everything Claude Code can do, available programmatically.
Google ADK
Google's official Agent Development Kit. Code-first, multi-language (Python/Go/TS/Java). Optimized for Gemini but model-agnostic. Built-in evaluation and deployment tools.
Vercel AI SDK
Provider-agnostic TypeScript toolkit from the Next.js creators. 20M+ monthly downloads. Works with React, Svelte, Vue, Angular. The standard for AI-powered web apps.
Pydantic AI
Type-safe agent framework built on Pydantic. Leverages Python's type system to catch agent logic errors at dev time. The Pythonic way to build AI agents.
Semantic Kernel
Microsoft's enterprise-grade AI orchestration SDK. Multi-language (C#/Python/Java). Core of the Microsoft Agent Framework. Strong Azure and Copilot integration.
Dify
Largest open-source agent platform. Production-ready visual orchestration for agentic workflows, RAG pipelines, and LLM apps. Visual builder plus code-level control.
OpenHands
Flagship open coding-agent framework (formerly OpenDevin). Plans, edits code, runs commands, and browses inside a sandbox. Frequently benchmarked on SWE-bench.
AutoGen
Microsoft's programming framework for agentic AI. Conversable, composable multi-agent patterns. Standard reference alongside LangGraph and CrewAI.
MetaGPT
Multi-agent framework that simulates a software company by assigning roles (PM, architect, engineer) to LLM agents. Canonical role-based pattern.
Mastra
TypeScript framework for AI applications and agents from the team behind Gatsby. Workflows, evals, and RAG with TS-native ergonomics.
AI Infrastructure
4Model gateways, observability, evals, and app-builder platforms
LiteLLM
Python SDK and proxy server (AI Gateway) that calls 100+ LLM APIs in OpenAI format. Cost tracking, guardrails, load balancing, logging. Powers ClaudePedia in production.
Langfuse
Open-source LLM engineering platform. Observability, metrics, evals, prompt management, playground, datasets. The clear category leader for LLM observability.
Phoenix
Arize's open-source AI observability and evaluation platform. OTel-native, LangChain instrumentation, ClaudePedia uses it for RAG tracing.
Promptfoo
Test prompts, agents, and RAGs declaratively. Red-teaming, vulnerability scanning, model comparison via CI/CD. The closest thing to 'pytest for LLMs'.
Tools
9Standalone CLI tools and utilities that enhance AI coding agents
RTK
Rust Token Killer. CLI proxy that rewrites dev commands for 60-90% token savings. Transparent hook-based usage, zero friction.
ccusage
Analyze Claude Code token usage and costs from local JSONL logs. Daily, monthly, and session reports with per-model breakdowns and 5-hour billing window tracking.
ccstatusline
Customizable status line for the Claude Code CLI. Displays model, git branch, token usage, and session metrics with Powerline themes.
Repomix
Packs an entire repository into a single AI-friendly file for LLM context windows. Tree-sitter compression (~70% token reduction), token counting, and MCP server.
CMUX
Native macOS terminal with vertical tabs, split panes, and notifications designed for running multiple AI coding agents in parallel.
Warp
AI-powered terminal built in Rust with GPU acceleration. Command suggestions, agentic workflows, and 700K+ developers. The terminal reimagined for AI.
Ghostty
Fast, cross-platform terminal by Mitchell Hashimoto (HashiCorp). Written in Zig with platform-native UI and GPU acceleration. 50K stars.
WezTerm
GPU-accelerated cross-platform terminal emulator and multiplexer in Rust. Popular with AI-CLI users for ligature support, multiplexing, and config-as-code via Lua.
claude-code-router
Drop-in router that lets the Claude Code CLI talk to any provider (OpenRouter, DeepSeek, Ollama, Qwen). Direct extension to Claude Code itself.
AI Design Tools
3Design-to-code tools and AI-powered visual editors
Pencil
AI-native vector design tool with MCP integration. Design-as-Code: turn prompts into UI directly on the live canvas.
Google Stitch
Google's AI design tool. Connects to your UI designs and generates production-ready code.
Figma
The collaborative design platform. AI features, dev mode, and MCP server support make it the bridge between design and code.
Skills & Plugins
7Claude Code skills, plugins, and CLI extensions we use daily
Superpowers
Agentic skills framework and software development methodology for Claude Code. Composable skills with built-in instructions that enforce discipline. 181K stars in 6 months — the breakout Claude Code plugin of 2026.
Get Shit Done (GSD)
Meta-prompting, context engineering, and spec-driven development for Claude Code. The workflow system we use to build ClaudePedia.
Compound Engineering
Office Compound Engineering plugin for Claude Code, Codex, and more. Structured workflows for engineering teams.
gstack
Full skill suite for Claude Code: QA, shipping, browsing, design review, investigation, deployment, and more. The backbone of our workflow.
UI/UX Pro Max
Design intelligence plugin. 50+ styles, 161 palettes, 57 font pairings, 99 UX guidelines. Searchable database for design decisions.
Context7
Fetches current library documentation on demand. No more stale training data for React, Next.js, Tailwind, or any framework.
Playwright CLI
Browser automation skill for testing and screenshots. Headless Chromium with persistent state across commands.
Last curated: May 2026
Go Deeper
The ecosystem is massive. These 120 aggregators do the heavy lifting: registries, benchmarks, curated lists, and landscape maps maintained by active communities.
Last curated: April 2026