# getdev.ai — tools for AI-assisted development

> Curated catalogue: coding agents, MCP, skills, agent frameworks, models, evals, and devops. Every entry is an established, widely-used project from a known maintainer. Hand-picked, not sponsored, no affiliate links.

Source: https://getdev.ai/tools · JSON: https://getdev.ai/tools.json
Last curated: 2026-07-12 · 103 entries across 15 categories

## Build with AI

Agents, editors, and builders that write code.

### Coding agents & CLIs

Agents that live in your terminal.

- [Claude Code](https://claude.com/claude-code) — Terminal agent that reads, edits, and ships whole repos. (by Anthropic) · [github](https://github.com/anthropics/claude-code) · [docs](https://docs.claude.com/en/docs/claude-code/overview)
- [Codex CLI](https://openai.com/codex/) — Open-source terminal agent wired to OpenAI models. (by OpenAI) · [github](https://github.com/openai/codex)
- [Gemini CLI](https://github.com/google-gemini/gemini-cli) — Open-source terminal agent with a generous free Gemini quota. (by Google) · [docs](https://google-gemini.github.io/gemini-cli/)
- [GitHub Copilot](https://github.com/features/copilot) — Completions, chat, and an async coding agent inside GitHub. (by GitHub) · [docs](https://docs.github.com/copilot)
- [Aider](https://aider.chat) — Open-source pair programmer with git-native diffs. (by Aider AI) · [github](https://github.com/Aider-AI/aider) · [docs](https://aider.chat/docs/)
- [Goose](https://block.github.io/goose/) — Open-source local agent, model-agnostic, MCP-native. (by Block) · [github](https://github.com/block/goose)
- [OpenCode](https://opencode.ai) — Open-source terminal agent; bring any model. (by SST) · [github](https://github.com/sst/opencode) · [docs](https://opencode.ai/docs)
- [Devin](https://devin.ai) — Autonomous software engineer you brief like a teammate. (by Cognition) · [docs](https://docs.devin.ai)

### AI editors & IDEs

Where the code gets written now.

- [Cursor](https://cursor.com) — VS Code fork with agent mode and repo-wide context. (by Anysphere) · [docs](https://docs.cursor.com)
- [Windsurf](https://windsurf.com) — Agentic IDE built around multi-file Cascade flows. (by Windsurf) · [docs](https://docs.windsurf.com)
- [Zed](https://zed.dev) — Fast native editor with agentic editing built in. (by Zed Industries) · [github](https://github.com/zed-industries/zed) · [docs](https://zed.dev/docs)
- [VS Code](https://code.visualstudio.com) — The default editor; Copilot and MCP built in. (by Microsoft) · [github](https://github.com/microsoft/vscode) · [docs](https://code.visualstudio.com/docs)
- [JetBrains AI](https://www.jetbrains.com/ai/) — AI assistant and Junie agent across the JetBrains IDEs. (by JetBrains)
- [Kiro](https://kiro.dev) — Spec-driven agentic IDE from AWS. (by AWS) · [docs](https://kiro.dev/docs/)

### Prompt-to-app builders

From a sentence to a deployed app.

- [v0](https://v0.app) — Prompt to production React and Next.js UI. (by Vercel)
- [Bolt](https://bolt.new) — Prompt, run, and deploy full-stack apps in the browser. (by StackBlitz) · [github](https://github.com/stackblitz/bolt.new)
- [Lovable](https://lovable.dev) — Chat your way to a deployed web app. (by Lovable) · [docs](https://docs.lovable.dev)
- [Replit](https://replit.com) — Browser IDE with an agent that builds and hosts. (by Replit) · [docs](https://docs.replit.com)
- [Firebase Studio](https://firebase.studio) — Agentic workspace on Gemini; deploys to Firebase. (by Google) · [docs](https://firebase.google.com/docs/studio)

## Extend & orchestrate

Protocols, skills, and frameworks around the model.

### Model Context Protocol

Wire tools and data into any agent.

- [MCP specification](https://modelcontextprotocol.io) — The open standard for connecting tools to agents. (by Anthropic) · [github](https://github.com/modelcontextprotocol/modelcontextprotocol)
- [MCP reference servers](https://github.com/modelcontextprotocol/servers) — Official servers: filesystem, git, fetch, memory. (by MCP project)
- [MCP Registry](https://registry.modelcontextprotocol.io) _(catalogue)_ — The official registry of published MCP servers. (by MCP project) · [github](https://github.com/modelcontextprotocol/registry)
- [MCP Inspector](https://github.com/modelcontextprotocol/inspector) — Debug UI for testing MCP servers locally. (by MCP project)
- [FastMCP](https://gofastmcp.com) — Pythonic framework for building MCP servers and clients. (by Prefect) · [github](https://github.com/jlowin/fastmcp)
- [Docker MCP Catalog](https://hub.docker.com/mcp) _(catalogue)_ — Containerised MCP servers, verified and signed by Docker. (by Docker) · [docs](https://docs.docker.com/ai/mcp-catalog-and-toolkit/)
- [Smithery](https://smithery.ai) _(catalogue)_ — Registry and hosting for thousands of MCP servers. (by Smithery)
- [awesome-mcp-servers](https://github.com/punkpeye/awesome-mcp-servers) _(catalogue)_ — The community's main curated list of MCP servers. (by community)

### Skills & agent standards

Conventions your repo can adopt today.

- [Agent Skills](https://github.com/anthropics/skills) — Reusable skill folders that teach agents procedures. (by Anthropic) · [docs](https://docs.claude.com/en/docs/claude-code/skills)
- [Claude Code plugins](https://docs.claude.com/en/docs/claude-code/plugins) — Package commands, agents, skills, and MCP servers for install. (by Anthropic)
- [AGENTS.md](https://agents.md) — One README-style file any coding agent can read. (by OpenAI + community) · [github](https://github.com/openai/agents.md)
- [llms.txt](https://llmstxt.org) — Proposed standard for LLM-readable site indexes. (by Answer.AI) · [github](https://github.com/AnswerDotAI/llms-txt)

### Agent frameworks & SDKs

Build your own agents.

- [Claude Agent SDK](https://github.com/anthropics/claude-agent-sdk-typescript) — Build custom agents on the Claude Code harness. (by Anthropic) · [docs](https://docs.claude.com/en/api/agent-sdk/overview)
- [OpenAI Agents SDK](https://github.com/openai/openai-agents-python) — Lightweight multi-agent orchestration in Python and TypeScript. (by OpenAI) · [docs](https://openai.github.io/openai-agents-python/)
- [Vercel AI SDK](https://ai-sdk.dev) — Provider-neutral TypeScript toolkit for AI apps and agents. (by Vercel) · [github](https://github.com/vercel/ai) · [docs](https://ai-sdk.dev/docs)
- [LangGraph](https://www.langchain.com/langgraph) — Graph-based orchestration for stateful, long-running agents. (by LangChain) · [github](https://github.com/langchain-ai/langgraph)
- [Pydantic AI](https://ai.pydantic.dev) — Type-safe agent framework from the Pydantic team. (by Pydantic) · [github](https://github.com/pydantic/pydantic-ai)
- [Mastra](https://mastra.ai) — TypeScript agent framework with workflows, RAG, and evals. (by Mastra) · [github](https://github.com/mastra-ai/mastra) · [docs](https://mastra.ai/docs)
- [CrewAI](https://www.crewai.com) — Role-based multi-agent crews in Python. (by CrewAI) · [github](https://github.com/crewAIInc/crewAI) · [docs](https://docs.crewai.com)
- [Agent Development Kit](https://github.com/google/adk-python) — The agent framework behind Google's own agent products. (by Google) · [docs](https://google.github.io/adk-docs/)
- [Agent Framework](https://github.com/microsoft/agent-framework) — Successor to AutoGen and Semantic Kernel. (by Microsoft) · [docs](https://learn.microsoft.com/en-us/agent-framework/)
- [DSPy](https://dspy.ai) — Program prompts declaratively; let the compiler optimise them. (by Stanford NLP) · [github](https://github.com/stanfordnlp/dspy)

## Models & inference

Hosted APIs and local runtimes.

### Model APIs & gateways

Hosted models behind an API key.

- [Anthropic API](https://www.anthropic.com/api) — Claude models for prose, code, and agents. (by Anthropic) · [docs](https://docs.claude.com)
- [OpenAI Platform](https://platform.openai.com) — GPT models, embeddings, images, and realtime voice. (by OpenAI) · [docs](https://platform.openai.com/docs)
- [Google AI Studio](https://aistudio.google.com) — Gemini API keys, playground, and a free tier. (by Google) · [docs](https://ai.google.dev)
- [Mistral](https://mistral.ai) — Open-weight and hosted models from Europe. (by Mistral AI) · [docs](https://docs.mistral.ai)
- [Groq](https://groq.com) — Open models served at very high speed on LPUs. (by Groq) · [docs](https://console.groq.com/docs)
- [OpenRouter](https://openrouter.ai) — One API key for hundreds of models. (by OpenRouter) · [docs](https://openrouter.ai/docs)
- [Together AI](https://www.together.ai) — Hosted open-source models, fine-tuning, GPU clusters. (by Together AI) · [docs](https://docs.together.ai)
- [Hugging Face](https://huggingface.co) — The hub for open models, datasets, and spaces. (by Hugging Face) · [docs](https://huggingface.co/docs)
- [Amazon Bedrock](https://aws.amazon.com/bedrock/) — Claude and other frontier models inside AWS. (by AWS) · [docs](https://docs.aws.amazon.com/bedrock/)
- [Vertex AI](https://cloud.google.com/vertex-ai) — Gemini, Claude, and open models inside Google Cloud. (by Google Cloud) · [docs](https://cloud.google.com/vertex-ai/docs)

### Local & open models

Models on your own hardware.

- [Ollama](https://ollama.com) — Run open models locally with one command. (by Ollama) · [github](https://github.com/ollama/ollama)
- [LM Studio](https://lmstudio.ai) — Desktop app for local models with an OpenAI-style API. (by LM Studio) · [docs](https://lmstudio.ai/docs)
- [vLLM](https://github.com/vllm-project/vllm) — High-throughput inference server for self-hosted models. (by vLLM project) · [docs](https://docs.vllm.ai)
- [llama.cpp](https://github.com/ggml-org/llama.cpp) — Run LLMs on CPUs and consumer hardware. (by ggml)
- [MLX](https://github.com/ml-explore/mlx) — Apple-silicon framework for local training and inference. (by Apple) · [docs](https://ml-explore.github.io/mlx/)

## Quality & operations

Evals, review, and infrastructure — the devops side.

### Evals & LLM observability

Know when the model is wrong.

- [LangSmith](https://www.langchain.com/langsmith) — Tracing, evals, and prompt management for LLM apps. (by LangChain) · [docs](https://docs.langchain.com)
- [Langfuse](https://langfuse.com) — Open-source LLM tracing, evals, and prompt management. (by Langfuse) · [github](https://github.com/langfuse/langfuse) · [docs](https://langfuse.com/docs)
- [Braintrust](https://www.braintrust.dev) — Eval-first platform for shipping reliable AI features. (by Braintrust) · [docs](https://www.braintrust.dev/docs)
- [promptfoo](https://www.promptfoo.dev) — Open-source prompt testing and red-teaming in CI. (by promptfoo) · [github](https://github.com/promptfoo/promptfoo)
- [Weave](https://wandb.ai/site/weave/) — Trace and evaluate LLM apps from the W&B stack. (by Weights & Biases) · [docs](https://weave-docs.wandb.ai)
- [Arize Phoenix](https://github.com/Arize-ai/phoenix) — Open-source tracing and evals built on OpenTelemetry. (by Arize AI) · [docs](https://arize.com/docs/phoenix)
- [Helicone](https://www.helicone.ai) — Open-source LLM gateway with logging and cost tracking. (by Helicone) · [github](https://github.com/Helicone/helicone)

### AI code review & CI

AI in the merge path.

- [CodeRabbit](https://www.coderabbit.ai) — AI review on every pull request. (by CodeRabbit) · [docs](https://docs.coderabbit.ai)
- [Claude Code GitHub Action](https://github.com/anthropics/claude-code-action) — Run Claude Code inside GitHub Actions on mention. (by Anthropic) · [docs](https://docs.claude.com/en/docs/claude-code/github-actions)
- [Qodo](https://www.qodo.ai) — Agentic code review and test generation. (by Qodo) · [docs](https://docs.qodo.ai)
- [Graphite](https://graphite.dev) — Stacked pull requests plus AI review. (by Graphite) · [docs](https://graphite.dev/docs)

### Agent infra & GPU compute

Sandboxes and GPUs for agent workloads.

- [E2B](https://e2b.dev) — Open-source sandboxes where agents run untrusted code. (by E2B) · [github](https://github.com/e2b-dev/E2B) · [docs](https://e2b.dev/docs)
- [Modal](https://modal.com) — Serverless GPU compute defined in Python. (by Modal) · [docs](https://modal.com/docs)
- [Replicate](https://replicate.com) — Hosted open-source models behind a single API. (by Replicate) · [docs](https://replicate.com/docs)
- [Baseten](https://www.baseten.co) — Dedicated model inference with fast cold starts. (by Baseten) · [docs](https://docs.baseten.co)
- [Workers AI](https://developers.cloudflare.com/workers-ai/) — Run models on Cloudflare's edge GPUs. (by Cloudflare)

### Ship & run

Where the shipped thing actually runs.

- [Vercel](https://vercel.com) — Serverless Next.js, Edge runtime, instant rollback. (by Vercel) · [docs](https://vercel.com/docs)
- [Railway](https://railway.app) — Container deploys with managed Postgres and Redis. (by Railway) · [docs](https://docs.railway.com)
- [Fly.io](https://fly.io) — Run containers close to users in many regions. (by Fly.io) · [docs](https://fly.io/docs)
- [Render](https://render.com) — Containers, managed Postgres, cron jobs in one panel. (by Render) · [docs](https://render.com/docs)
- [Cloudflare](https://www.cloudflare.com) — CDN, Workers, R2, Pages — the edge stack. (by Cloudflare) · [docs](https://developers.cloudflare.com)
- [Supabase](https://supabase.com) — Managed Postgres with auth, storage, realtime hooks. (by Supabase) · [github](https://github.com/supabase/supabase) · [docs](https://supabase.com/docs)
- [Neon](https://neon.com) — Serverless Postgres with branch databases for previews. (by Neon) · [docs](https://neon.com/docs)
- [Upstash](https://upstash.com) — Serverless Redis, queues, and vector storage. (by Upstash) · [docs](https://upstash.com/docs)
- [Sentry](https://sentry.io) — Error tracking and performance monitoring. (by Sentry) · [github](https://github.com/getsentry/sentry) · [docs](https://docs.sentry.io)
- [Axiom](https://axiom.co) — Logs and traces with SQL-shaped queries. (by Axiom) · [docs](https://axiom.co/docs)
- [PostHog](https://posthog.com) — Product analytics with session replay and feature flags. (by PostHog) · [github](https://github.com/PostHog/posthog) · [docs](https://posthog.com/docs)

## Catalogues & reference

Trusted places to find more.

### Curated awesome lists

Community-vetted link lists, maintained in the open.

- [awesome](https://github.com/sindresorhus/awesome) _(catalogue)_ — The index of awesome lists on every topic. (by Sindre Sorhus)
- [awesome-rust](https://github.com/rust-unofficial/awesome-rust) _(catalogue)_ — Curated Rust crates and resources. (by community)
- [awesome-go](https://awesome-go.com) _(catalogue)_ — Curated Go packages, vetted by contribution rules. (by community) · [github](https://github.com/avelino/awesome-go)
- [awesome-python](https://github.com/vinta/awesome-python) _(catalogue)_ — Curated Python frameworks, libraries, and tooling. (by community)
- [awesome-selfhosted](https://awesome-selfhosted.net) _(catalogue)_ — Self-hostable alternatives to SaaS, categorised. (by community) · [github](https://github.com/awesome-selfhosted/awesome-selfhosted)
- [awesome-ai-agents](https://github.com/e2b-dev/awesome-ai-agents) _(catalogue)_ — Curated catalogue of AI agent projects and frameworks. (by E2B)
- [awesome-llm-apps](https://github.com/Shubhamsaboo/awesome-llm-apps) _(catalogue)_ — LLM app examples with RAG and agent patterns. (by Shubham Saboo)
- [awesome-claude-code](https://github.com/hesreallyhim/awesome-claude-code) _(catalogue)_ — Curated Claude Code commands, skills, and workflows. (by community)

### Official package registries

The canonical sources; check publishers before installing.

- [npm](https://www.npmjs.com) _(catalogue)_ — The JavaScript package registry. (by GitHub)
- [PyPI](https://pypi.org) _(catalogue)_ — The Python package index. (by Python Software Foundation)
- [crates.io](https://crates.io) _(catalogue)_ — The Rust package registry. (by Rust Foundation)
- [pkg.go.dev](https://pkg.go.dev) _(catalogue)_ — Go package discovery and docs. (by Google)
- [Docker Hub](https://hub.docker.com) _(catalogue)_ — Container images; prefer official and verified publishers. (by Docker)

### Docs & learning

Primary sources worth the reading time.

- [Claude docs](https://docs.claude.com) — API reference, prompt guides, Claude Code manual. (by Anthropic)
- [Anthropic Cookbook](https://github.com/anthropics/anthropic-cookbook) — Notebooks of working Claude patterns. (by Anthropic)
- [OpenAI Cookbook](https://cookbook.openai.com) — Recipes and guides for the OpenAI API. (by OpenAI) · [github](https://github.com/openai/openai-cookbook)
- [Prompt Engineering Guide](https://www.promptingguide.ai) — The standard reference for prompting techniques. (by DAIR.AI) · [github](https://github.com/dair-ai/Prompt-Engineering-Guide)
- [Hugging Face Learn](https://huggingface.co/learn) — Free courses: LLMs, agents, deep RL, diffusion. (by Hugging Face)
- [DeepLearning.AI](https://www.deeplearning.ai) — Short courses on agents, evals, and LLM engineering. (by DeepLearning.AI)
- [Simon Willison's weblog](https://simonwillison.net) — Daily field notes on LLMs and agentic coding. (by Simon Willison)