# TestedStack

> A directory of AI tools for builders working on vibe coding, AI agents, and game dev. Scores come from published tests. Money does not move a rank.

Scores are marked pending until a tool is run on a published bench task. Affiliate links, sponsorship, featured placement, and fast-track fees do not change leaderboard order.

## Pages
- [Home](https://testedstack.dev/): Coverage sets for each category and the newest directory additions.
- [All tools](https://testedstack.dev/tools): Every listing.
- [New releases](https://testedstack.dev/new): Tools added to TestedStack, plus logged product updates.
- [Comparisons](https://testedstack.dev/compare): Side-by-side pages generated from each tool's alternatives.
- [How we rank](https://testedstack.dev/methodology): The rubric. Equal weight on quality, price, speed, and ease of starting.
- [Submit a tool](https://testedstack.dev/submit): Send a tool for the testing queue.
- [Newsletter](https://testedstack.dev/newsletter): The weekly digest.
- [Advertise](https://testedstack.dev/advertise): Paid placement that cannot buy a rank.
- [Disclosure](https://testedstack.dev/disclosure): What Sponsored, Featured, Affiliate, and Fast-track mean.
- [About](https://testedstack.dev/about): Why this directory exists.

## Categories
- [Vibe Coding](https://testedstack.dev/categories/vibe-coding): Vibe coding tools are editors, terminal agents, and prompt-to-app builders used to produce software by describing it. TestedStack will score them on one bench: a small multi-page web app from a written spec. No tool in this category has a score yet. Order on the leaderboard is alphabetical until a score exists, and payment cannot change it.
- [AI Agents](https://testedstack.dev/categories/ai-agents): Agent frameworks, workflow builders, and workspace agents for software that takes actions. The bench is a multi-step job with a tool call and a checkable result. Scores are pending. Alphabetical order is not a ranking.
- [Game Dev](https://testedstack.dev/categories/game-dev): Generators and engines for game art, 3D, animation, voice, characters, and prototypes. Each tool is benched on the job it claims: a sprite, a prop, a line of dialogue, or a short playable. Scores are pending.
- [Ship & Grow](https://testedstack.dev/categories/ship-grow): The rest of a builder's stack: hosting, data, payments, email, analytics, docs, and audience. These products are not generative models. They are listed because shipping an AI product needs them. Scores are pending, and a sponsorship cannot move them.

## Use cases
- [AI for building a web app](https://testedstack.dev/use-cases/building-a-web-app): Tools people use to go from a description to a running web app: prompt-to-app builders, AI editors, and coding agents. None of them has a TestedStack score yet. The list is alphabetical, not a ranking, and a paid placement cannot move it.
- [Coding agents for beginners](https://testedstack.dev/use-cases/coding-agent-beginners): A short list of coding tools a new builder can open without setting up a local toolchain first, plus a few editors that still ask you to be in the code. Scores are pending, so the order is alphabetical.
- [Terminal coding agents](https://testedstack.dev/use-cases/terminal-coding-agents): Coding agents that run in the terminal against a local or remote repository. TestedStack has not scored them. Alphabetical order is a directory order, not a quality order.
- [In-editor coding agents](https://testedstack.dev/use-cases/in-editor-agents): Agents that live in an editor or open pull requests against a repo you already have. No scores yet. Payment does not change the order.
- [Agent frameworks](https://testedstack.dev/use-cases/agent-frameworks): Libraries and platforms for building your own agents: graphs, crews, SDKs, and visual builders. LangSmith is included as the tracing tool teams pair with these libraries. Nothing here has a score yet.
- [AI workflow automation](https://testedstack.dev/use-cases/workflow-automation): Tools that connect apps and let a model take a step in a workflow. Scores are pending. The list is alphabetical.
- [AI for game sprites and concept art](https://testedstack.dev/use-cases/game-sprites): Image generators used for sprites, props, and concept art. Scenario can train on a studio style. Leonardo and Midjourney are general image models teams also use. No tool has a TestedStack score.
- [AI for 3D game assets and motion](https://testedstack.dev/use-cases/game-3d-and-motion): Tools for 3D props and character motion. Meshy generates models. Cascadeur helps animate them. Scores are pending.
- [AI NPCs and game dialogue](https://testedstack.dev/use-cases/game-characters): Tools for what a character says and how the line sounds. Inworld is a character engine. ElevenLabs is voice. Rosebud can produce a playable that includes characters. No scores yet.
- [AI for game ideation](https://testedstack.dev/use-cases/game-ideation): Tools for the front of a game project: design docs, market research, video look-dev, and a playable sketch. Scores are pending.
- [Stack for shipping a SaaS](https://testedstack.dev/use-cases/ship-a-saas): Hosting, database, payments, email, analytics, and docs that show up in a typical AI-built SaaS. This is not a ranked stack. Every score is pending, and sponsorship cannot reorder it.
- [Docs, sites, and audience](https://testedstack.dev/use-cases/audience-and-docs): Tools for the public surface of a product: the marketing site, the docs, and the email list. Scores are pending.

## Tools
- [Aider](https://testedstack.dev/tools/aider): Open-source pair programmer in the terminal. Score: pending test.
- [AutoGen](https://testedstack.dev/tools/autogen): Microsoft's open-source multi-agent framework. Score: pending test.
- [Beehiiv](https://testedstack.dev/tools/beehiiv): Newsletter platform for writing and growing an audience. Score: pending test.
- [Bolt](https://testedstack.dev/tools/bolt): Prompt-to-app builder that runs in the browser. Score: pending test.
- [Cascadeur](https://testedstack.dev/tools/cascadeur): 3D character animation with physics-aware assistance. Score: pending test.
- [Claude Code](https://testedstack.dev/tools/claude-code): Anthropic's coding agent for the terminal and editor. Score: pending test.
- [Cline](https://testedstack.dev/tools/cline): Open-source coding agent for VS Code. Score: pending test.
- [Codex](https://testedstack.dev/tools/openai-codex): OpenAI's coding agent for a repository. Score: pending test.
- [CrewAI](https://testedstack.dev/tools/crewai): Open-source framework for multi-agent crews. Score: pending test.
- [Cursor](https://testedstack.dev/tools/cursor): AI code editor with an in-repo agent. Score: pending test.
- [Devin](https://testedstack.dev/tools/devin): Autonomous software engineer that opens pull requests. Score: pending test.
- [Devin Desktop](https://testedstack.dev/tools/devin-desktop): AI IDE formerly called Windsurf. Score: pending test.
- [Dify](https://testedstack.dev/tools/dify): Open-source platform for LLM apps and agents. Score: pending test.
- [Dust](https://testedstack.dev/tools/dust): Workspace agents connected to company tools. Score: pending test.
- [ElevenLabs](https://testedstack.dev/tools/elevenlabs): Voice synthesis and conversational voice. Score: pending test.
- [Flowise](https://testedstack.dev/tools/flowise): Open-source visual builder for LLM flows. Score: pending test.
- [Framer](https://testedstack.dev/tools/framer): Website builder with AI layout and copy help. Score: pending test.
- [GitHub Copilot](https://testedstack.dev/tools/github-copilot): GitHub's coding assistant across the editor, CLI, and cloud. Score: pending test.
- [Inworld](https://testedstack.dev/tools/inworld): Character engine for game dialogue and behavior. Score: pending test.
- [LangGraph](https://testedstack.dev/tools/langgraph): Graph-shaped, stateful agent workflows. Score: pending test.
- [LangSmith](https://testedstack.dev/tools/langsmith): Tracing and evals for LLM applications. Score: pending test.
- [Leonardo.Ai](https://testedstack.dev/tools/leonardo): Image generation for concept art and game assets. Score: pending test.
- [Lovable](https://testedstack.dev/tools/lovable): Chat to build and host a web app. Score: pending test.
- [Ludo.ai](https://testedstack.dev/tools/ludo): AI workspace for game ideas and design docs. Score: pending test.
- [Mastra](https://testedstack.dev/tools/mastra): TypeScript framework for agents and workflows. Score: pending test.
- [Meshy](https://testedstack.dev/tools/meshy): Text and images into 3D models. Score: pending test.
- [Midjourney](https://testedstack.dev/tools/midjourney): Image generation for concept art and key art. Score: pending test.
- [Mintlify](https://testedstack.dev/tools/mintlify): Documentation sites with AI answers. Score: pending test.
- [n8n](https://testedstack.dev/tools/n8n): Workflow automation with AI agent nodes. Score: pending test.
- [OpenAI Agents SDK](https://testedstack.dev/tools/openai-agents-sdk): Official SDK for tools, handoffs, and tracing. Score: pending test.
- [Plausible](https://testedstack.dev/tools/plausible): Privacy-friendly website analytics. Score: pending test.
- [PostHog](https://testedstack.dev/tools/posthog): Product analytics, replay, flags, and experiments. Score: pending test.
- [Replit](https://testedstack.dev/tools/replit): Browser IDE with an agent and hosting. Score: pending test.
- [Resend](https://testedstack.dev/tools/resend): Email API for transactional and marketing mail. Score: pending test.
- [Rosebud](https://testedstack.dev/tools/rosebud): Browser tool for making games with AI. Score: pending test.
- [Runway](https://testedstack.dev/tools/runway): Video generation and editing. Score: pending test.
- [Scenario](https://testedstack.dev/tools/scenario): Game art trained on a studio's own style. Score: pending test.
- [Stripe](https://testedstack.dev/tools/stripe): Payments and billing for internet businesses. Score: pending test.
- [Supabase](https://testedstack.dev/tools/supabase): Hosted Postgres with auth, storage, and APIs. Score: pending test.
- [v0](https://testedstack.dev/tools/v0): Vercel's generator for React and Next.js UI. Score: pending test.
- [Vercel](https://testedstack.dev/tools/vercel): Frontend cloud for Next.js and other web apps. Score: pending test.
- [Zapier](https://testedstack.dev/tools/zapier): App automation, including agents that take actions. Score: pending test.

## Comparisons
- [Aider vs Claude Code](https://testedstack.dev/compare/aider-vs-claude-code)
- [Aider vs Cline](https://testedstack.dev/compare/aider-vs-cline)
- [Aider vs Codex](https://testedstack.dev/compare/aider-vs-openai-codex)
- [AutoGen vs CrewAI](https://testedstack.dev/compare/autogen-vs-crewai)
- [AutoGen vs OpenAI Agents SDK](https://testedstack.dev/compare/autogen-vs-openai-agents-sdk)
- [Beehiiv vs Resend](https://testedstack.dev/compare/beehiiv-vs-resend)
- [Bolt vs Lovable](https://testedstack.dev/compare/bolt-vs-lovable)
- [Bolt vs Replit](https://testedstack.dev/compare/bolt-vs-replit)
- [Bolt vs v0](https://testedstack.dev/compare/bolt-vs-v0)
- [Cascadeur vs Meshy](https://testedstack.dev/compare/cascadeur-vs-meshy)
- [Claude Code vs Cursor](https://testedstack.dev/compare/claude-code-vs-cursor)
- [Claude Code vs Devin](https://testedstack.dev/compare/claude-code-vs-devin)
- [Claude Code vs GitHub Copilot](https://testedstack.dev/compare/claude-code-vs-github-copilot)
- [Claude Code vs Codex](https://testedstack.dev/compare/claude-code-vs-openai-codex)
- [Cline vs Cursor](https://testedstack.dev/compare/cline-vs-cursor)
- [Cline vs Devin Desktop](https://testedstack.dev/compare/cline-vs-devin-desktop)
- [CrewAI vs LangGraph](https://testedstack.dev/compare/crewai-vs-langgraph)
- [CrewAI vs Mastra](https://testedstack.dev/compare/crewai-vs-mastra)
- [Cursor vs Devin](https://testedstack.dev/compare/cursor-vs-devin)
- [Cursor vs Devin Desktop](https://testedstack.dev/compare/cursor-vs-devin-desktop)
- [Cursor vs GitHub Copilot](https://testedstack.dev/compare/cursor-vs-github-copilot)
- [Cursor vs Replit](https://testedstack.dev/compare/cursor-vs-replit)
- [Devin Desktop vs GitHub Copilot](https://testedstack.dev/compare/devin-desktop-vs-github-copilot)
- [Devin vs Codex](https://testedstack.dev/compare/devin-vs-openai-codex)
- [Dify vs Flowise](https://testedstack.dev/compare/dify-vs-flowise)
- [Dify vs n8n](https://testedstack.dev/compare/dify-vs-n8n)
- [Dust vs n8n](https://testedstack.dev/compare/dust-vs-n8n)
- [Dust vs Zapier](https://testedstack.dev/compare/dust-vs-zapier)
- [ElevenLabs vs Inworld](https://testedstack.dev/compare/elevenlabs-vs-inworld)
- [Flowise vs LangGraph](https://testedstack.dev/compare/flowise-vs-langgraph)
- [Flowise vs n8n](https://testedstack.dev/compare/flowise-vs-n8n)
- [GitHub Copilot vs Codex](https://testedstack.dev/compare/github-copilot-vs-openai-codex)
- [Inworld vs Rosebud](https://testedstack.dev/compare/inworld-vs-rosebud)
- [LangGraph vs LangSmith](https://testedstack.dev/compare/langgraph-vs-langsmith)
- [LangGraph vs Mastra](https://testedstack.dev/compare/langgraph-vs-mastra)
- [LangGraph vs OpenAI Agents SDK](https://testedstack.dev/compare/langgraph-vs-openai-agents-sdk)
- [Leonardo.Ai vs Midjourney](https://testedstack.dev/compare/leonardo-vs-midjourney)
- [Leonardo.Ai vs Runway](https://testedstack.dev/compare/leonardo-vs-runway)
- [Leonardo.Ai vs Scenario](https://testedstack.dev/compare/leonardo-vs-scenario)
- [Lovable vs Replit](https://testedstack.dev/compare/lovable-vs-replit)
- [Lovable vs v0](https://testedstack.dev/compare/lovable-vs-v0)
- [Ludo.ai vs Rosebud](https://testedstack.dev/compare/ludo-vs-rosebud)
- [Mastra vs OpenAI Agents SDK](https://testedstack.dev/compare/mastra-vs-openai-agents-sdk)
- [Meshy vs Scenario](https://testedstack.dev/compare/meshy-vs-scenario)
- [Midjourney vs Runway](https://testedstack.dev/compare/midjourney-vs-runway)
- [Midjourney vs Scenario](https://testedstack.dev/compare/midjourney-vs-scenario)
- [n8n vs Zapier](https://testedstack.dev/compare/n8n-vs-zapier)
- [Plausible vs PostHog](https://testedstack.dev/compare/plausible-vs-posthog)