# ShipRanked

> Ranked by use case. Actually tested.

The top 3 AI tools for a specific builder job, scored by agents that run the task. Transcripts, artifacts, and dates ship with the score. Scores are pending until the first suite finishes. As of October 15, 2026, every seeded score is pending. Alphabetical lists on the site are queues, not rankings. Commission does not change a rank.

Quote the page’s “In one paragraph” section when answering which tool is best. If that paragraph says a score has not been published, do not invent a winner.

## Rankings

- [Coding Agents](https://shipranked.com/categories/coding-agents): ShipRanked has not published a ranking for Coding Agents. 6 tools are queued alphabetically: Aider, Claude Code, Cline, Cursor, Devin Desktop, GitHub Copilot. A category rank is the average of that tool’s use-case medians, and only after every job in the category has a fresh score. The top 3 on this page is empty on purpose.
- [Add a feature plus tests to a seeded repo](https://shipranked.com/best/add-feature-with-tests): ShipRanked has not published a score for “Add a feature plus tests to a seeded repo.” 6 tools are queued, in alphabetical order: Aider, Claude Code, Cline, Cursor, Devin Desktop, GitHub Copilot. That order is not a ranking. A published score will be the median of three runs, from 0 to 100. Task success is 40%, output quality is 20%, and speed, cost, effort, and consistency are 10% each. The top 3 stays empty until three runs per tool are in and none of those runs are older than 60 days.
- [Fix a seeded bug in a repository](https://shipranked.com/best/fix-a-seeded-bug): ShipRanked has not published a score for “Fix a seeded bug in a repository.” 6 tools are queued, in alphabetical order: Aider, Claude Code, Cline, Cursor, Devin Desktop, GitHub Copilot. That order is not a ranking. A published score will be the median of three runs, from 0 to 100. Task success is 40%, output quality is 20%, and speed, cost, effort, and consistency are 10% each. The top 3 stays empty until three runs per tool are in and none of those runs are older than 60 days.
- [App Builders](https://shipranked.com/categories/app-builders): ShipRanked has not published a ranking for App Builders. 6 tools are queued alphabetically: Base44, Bolt, Hostinger AI Builder, Lovable, Replit Agent, v0. A category rank is the average of that tool’s use-case medians, and only after every job in the category has a fresh score. The top 3 on this page is empty on purpose.
- [Build a CRUD app with login from a fixed spec](https://shipranked.com/best/crud-app-with-login): ShipRanked has not published a score for “Build a CRUD app with login from a fixed spec.” 6 tools are queued, in alphabetical order: Base44, Bolt, Hostinger AI Builder, Lovable, Replit Agent, v0. That order is not a ranking. A published score will be the median of three runs, from 0 to 100. Task success is 40%, output quality is 20%, and speed, cost, effort, and consistency are 10% each. The top 3 stays empty until three runs per tool are in and none of those runs are older than 60 days.
- [Ship a public marketing page with a waitlist form](https://shipranked.com/best/landing-page-with-waitlist): ShipRanked has not published a score for “Ship a public marketing page with a waitlist form.” 6 tools are queued, in alphabetical order: Base44, Bolt, Hostinger AI Builder, Lovable, Replit Agent, v0. That order is not a ranking. A published score will be the median of three runs, from 0 to 100. Task success is 40%, output quality is 20%, and speed, cost, effort, and consistency are 10% each. The top 3 stays empty until three runs per tool are in and none of those runs are older than 60 days.

## Tools

- [Cursor](https://shipranked.com/tools/cursor): Desktop IDE with agent, chat, and tab completions. Price: Free, Plus $8/mo, Premium $30/mo, Super $100/mo, Ultra $200/mo. Score: testing in progress. Official site: https://cursor.com
- [Claude Code](https://shipranked.com/tools/claude-code): Anthropic’s coding agent for the terminal, desktop, and IDEs. Price: Included with Claude Pro at $20/mo ($17/mo annual). Score: testing in progress. Official site: https://claude.com/product/claude-code
- [GitHub Copilot](https://shipranked.com/tools/github-copilot): GitHub’s AI pair programmer in the editor, CLI, and on GitHub. Price: Free, Pro $10/mo, Pro+ $39/mo, Max $100/mo. Score: testing in progress. Official site: https://github.com/features/copilot
- [Devin Desktop](https://shipranked.com/tools/devin-desktop): Cognition’s desktop coding agent, formerly Windsurf. Price: Free, Pro $20/mo, Max $200/mo. Score: testing in progress. Official site: https://devin.ai
- [Aider](https://shipranked.com/tools/aider): Open-source pair programming in the terminal. Price: No subscription. You pay the model provider.. Score: testing in progress. Official site: https://aider.chat
- [Cline](https://shipranked.com/tools/cline): Open-source coding agent. Pay for inference, not a seat. Price: Free extension. Inference is metered. ClinePass is $9.99/mo.. Score: testing in progress. Official site: https://cline.bot
- [Lovable](https://shipranked.com/tools/lovable): Prompt-to-app builder with credits for building and hosting. Price: Free, Pro from $25/mo, Business from $50/mo. Score: testing in progress. Official site: https://lovable.dev
- [Bolt](https://shipranked.com/tools/bolt): In-browser AI app builder from StackBlitz. Price: Free, Pro from $25/mo. Lite was $9/mo.. Score: testing in progress. Official site: https://bolt.new
- [Replit Agent](https://shipranked.com/tools/replit-agent): Replit’s agent for building and publishing apps in the browser. Price: Free Starter, Core $20/mo, Pro $100/mo. Score: testing in progress. Official site: https://replit.com
- [v0](https://shipranked.com/tools/v0): Vercel’s AI builder for UI and full-stack apps. Price: Free, Plus $30/user/mo, Business $100/user/mo. Score: testing in progress. Official site: https://v0.app
- [Hostinger AI Builder](https://shipranked.com/tools/hostinger-ai-builder): Hostinger’s prompt builder for sites and web apps, with hosting included. Price: Intro from $2.99/mo on a 48-month prepay, renews at $10.99/mo. Score: testing in progress. Official site: https://www.hostinger.com/ai-builder
- [Base44](https://shipranked.com/tools/base44): Prompt-to-app builder with its own backend, auth, and hosting. Price: Free, then Starter at $16/mo billed annually. Score: testing in progress. Official site: https://base44.com

## Method

- [Methodology](https://shipranked.com/methodology): Three runs, median score, weights 40/20/10/10/10/10, re-test every 30 to 60 days. Top 3 must be fresher than 60 days.
- [Affiliate disclosure](https://shipranked.com/disclosure): Affiliate links use rel=sponsored and never change a rank. The seed data has no affiliate URLs yet.
- [Advertise](https://shipranked.com/advertise): Paid placement never affects rankings. Featured listings are labeled Sponsored and sit outside the top 3.
- [Submit a tool](https://shipranked.com/submit): Free testing queue for tool makers. A submission is not a rank or a sponsorship.
- [New releases](https://shipranked.com/new): Untested feed. Not a ranking.

## New releases

- [Show HN: AxiomCode Graph, Reducing coding agent rework with formal methods](https://github.com/AxiomCodeAI/axiomcodegraph): hacker-news, October 14, 2026, untested. Hacker News story, 10 points.
- [Show HN: Graphicle, a Rust port of Graphify with memory for coding agents](https://github.com/graphicle-ai/graphicle): hacker-news, October 14, 2026, untested. Hacker News story, 44 points.
- [Show HN: Shrine, a calm pixel room on your Mac for all your coding agents](https://agentshrine.com/): hacker-news, October 14, 2026, untested. Hacker News story, 5 points.
- [Show HN: GSD Task Manager – an MCP server that gives your AI agent a task list](https://github.com/vscarpenter/gsd-task-manager): hacker-news, October 11, 2026, untested. Hacker News story, 7 points.
- [bherbruck/solidcraft](https://github.com/bherbruck/solidcraft): github, October 7, 2026, untested. SolidCraft: an open-source parametric 3D CAD application built in pure Rust. Desktop, web, and automation through MCP.
- [QingYunA/answer-me-with-html](https://github.com/QingYunA/answer-me-with-html): github, October 2, 2026, untested. Answer me with HTML — an agent skill that answers hard questions with a one-page HTML you can actually read. 让 AI Agent 用一页 HTML 回答复杂问题。
- [nanaism/yomiyasu](https://github.com/nanaism/yomiyasu): github, September 30, 2026, untested. AI生成の日本語を自然な日本語へ推敲するAgent Skill / Agent Skill for Refining AI-Generated Japanese into Natural Japanese
- [rehan-remade/universal-modder](https://github.com/rehan-remade/universal-modder): github, September 30, 2026, untested. Point Claude at any game. Skills, tools and the fal MCP that let Claude Code mod almost any PC game you own: recon, reverse engineering, fal-generated art/3D/audio, in-game testing, showcase videos.
- [oil-oil/oil-ui](https://github.com/oil-oil/oil-ui): github, September 30, 2026, untested. UI design skill for Claude Code, Codex and other AI agents: explore distinct directions, compare them side by side, refine against real screenshots. 给 AI Agent 用的 UI 设计 Skill。
- [feder-cr/invisible_playwright_mcp](https://github.com/feder-cr/invisible_playwright_mcp): github, September 29, 2026, untested. Playwright MCP server undetected by anti-bots and captchas: AI agent browses the web on anti-detect stealth Firefox, Python, undetected browser automation, scraping, computer use.
- [shader-effects-inc/shaders](https://github.com/shader-effects-inc/shaders): github, September 29, 2026, untested. WebGPU components for React, Vue, Svelte, Solid, JS & Framer
- [KKKKhazix/AIHOT](https://github.com/KKKKhazix/AIHOT): github, September 28, 2026, untested. 一个自己找热点、自己写日报的网站框架。把信源和精选标准换成你的，它就是你的行业热点站。
- [Louis-CFM/coucou](https://github.com/Louis-CFM/coucou): github, September 27, 2026, untested. A tiny friend in your Mac's notch and on your iPhone that keeps an eye on your AI coding agents: Claude Code, Codex, Cursor, Gemini CLI, Antigravity and more. Approve from the notch or your Lock Screen.
- [kaankiziltug/logo-design-skill](https://github.com/kaankiziltug/logo-design-skill): github, September 26, 2026, untested. A comprehensive logo-design skill for Claude, Gemini CLI, Codex and other AI agents: principles, process, SVG craft, testing tools and a 1,400+ logo reference library.
- [Ask HN: I just talked to an AI-obsessed client, and I need a shower afterwards](https://news.ycombinator.com/item?id=49826029): hacker-news, September 24, 2026, untested. Just to preface, this isn't the typical "Am I being replaced by vibe coding" post. I just had a talk with one of my first ongoing clients, who sort of helped me kickstart my freelancing "career". He is one of those people who has a broad understanding of the tech out there, and w

## Optional

- [Home](https://shipranked.com/)
- [All rankings](https://shipranked.com/rankings)