Tools.

Terminal agent that reads, edits, and ships whole repos.

Open-source terminal agent wired to OpenAI models.

Open-source terminal agent with a generous free Gemini quota.

Completions, chat, and an async coding agent inside GitHub.

Open-source pair programmer with git-native diffs.

Open-source local agent, model-agnostic, MCP-native.

Open-source terminal agent; bring any model.

Autonomous software engineer you brief like a teammate.

VS Code fork with agent mode and repo-wide context.

Agentic IDE built around multi-file Cascade flows.

Fast native editor with agentic editing built in.

The default editor; Copilot and MCP built in.

AI assistant and Junie agent across the JetBrains IDEs.

Spec-driven agentic IDE from AWS.

Prompt to production React and Next.js UI.

Prompt, run, and deploy full-stack apps in the browser.

Chat your way to a deployed web app.

Browser IDE with an agent that builds and hosts.

Agentic workspace on Gemini; deploys to Firebase.

The open standard for connecting tools to agents.

Official servers: filesystem, git, fetch, memory.

The official registry of published MCP servers.

Debug UI for testing MCP servers locally.

Pythonic framework for building MCP servers and clients.

Containerised MCP servers, verified and signed by Docker.

Registry and hosting for thousands of MCP servers.

The community's main curated list of MCP servers.

Reusable skill folders that teach agents procedures.

Package commands, agents, skills, and MCP servers for install.

One README-style file any coding agent can read.

Proposed standard for LLM-readable site indexes.

Build custom agents on the Claude Code harness.

Lightweight multi-agent orchestration in Python and TypeScript.

Provider-neutral TypeScript toolkit for AI apps and agents.

Graph-based orchestration for stateful, long-running agents.

Type-safe agent framework from the Pydantic team.

TypeScript agent framework with workflows, RAG, and evals.

Role-based multi-agent crews in Python.

The agent framework behind Google's own agent products.

Successor to AutoGen and Semantic Kernel.

Program prompts declaratively; let the compiler optimise them.

Claude models for prose, code, and agents.

GPT models, embeddings, images, and realtime voice.

Gemini API keys, playground, and a free tier.

Open-weight and hosted models from Europe.

Open models served at very high speed on LPUs.

One API key for hundreds of models.

Hosted open-source models, fine-tuning, GPU clusters.

The hub for open models, datasets, and spaces.

Claude and other frontier models inside AWS.

Gemini, Claude, and open models inside Google Cloud.

Run open models locally with one command.

Desktop app for local models with an OpenAI-style API.

High-throughput inference server for self-hosted models.

Run LLMs on CPUs and consumer hardware.

Apple-silicon framework for local training and inference.

Tracing, evals, and prompt management for LLM apps.

Open-source LLM tracing, evals, and prompt management.

Eval-first platform for shipping reliable AI features.

Open-source prompt testing and red-teaming in CI.

Trace and evaluate LLM apps from the W&B stack.

Open-source tracing and evals built on OpenTelemetry.

Open-source LLM gateway with logging and cost tracking.

AI review on every pull request.

Run Claude Code inside GitHub Actions on mention.

Agentic code review and test generation.

Stacked pull requests plus AI review.

Open-source sandboxes where agents run untrusted code.

Serverless GPU compute defined in Python.

Hosted open-source models behind a single API.

Dedicated model inference with fast cold starts.

Run models on Cloudflare's edge GPUs.

Serverless Next.js, Edge runtime, instant rollback.

Container deploys with managed Postgres and Redis.

Run containers close to users in many regions.

Containers, managed Postgres, cron jobs in one panel.

CDN, Workers, R2, Pages — the edge stack.

Managed Postgres with auth, storage, realtime hooks.

Serverless Postgres with branch databases for previews.

Serverless Redis, queues, and vector storage.

Error tracking and performance monitoring.

Logs and traces with SQL-shaped queries.

Product analytics with session replay and feature flags.

The index of awesome lists on every topic.

Curated Rust crates and resources.

Curated Go packages, vetted by contribution rules.

Curated Python frameworks, libraries, and tooling.

Self-hostable alternatives to SaaS, categorised.

Curated catalogue of AI agent projects and frameworks.

LLM app examples with RAG and agent patterns.

Curated Claude Code commands, skills, and workflows.

The JavaScript package registry.

The Python package index.

The Rust package registry.

Go package discovery and docs.

Container images; prefer official and verified publishers.

API reference, prompt guides, Claude Code manual.

Notebooks of working Claude patterns.

Recipes and guides for the OpenAI API.

The standard reference for prompting techniques.

Free courses: LLMs, agents, deep RL, diffusion.

Short courses on agents, evals, and LLM engineering.

Daily field notes on LLMs and agentic coding.