The AI map · Building with AI

Coding agents

Agents that read a codebase, write and change code, run tests and open pull requests for a developer, in the terminal, the editor or the cloud. The ones that lead it as of 29 October 2026.

  1. Claude Code New since July

    Anthropic

    An agent for developers that plans, writes code, runs tests and opens pull requests, from the terminal, IDE, Slack, web or desktop app.

    Pick it Choose it for long refactors and migrations where you want the agent to keep going for hours, and you already pay for Claude.

    Price Included in all paid Claude plans: Pro $20 a month ($17 a month billed annually), Max from $100 a month

    Source: claude.com

  2. Codex New since July

    OpenAI

    OpenAI's coding agent, in ChatGPT, the editor and the terminal, that runs many agents in parallel and reviews pull requests.

    Pick it Choose it when you already pay for ChatGPT, or want tasks to keep running in the cloud after you close the laptop.

    Price Included in ChatGPT plans: limited on Free, more on Plus at $20 a month, Pro from $100 a month

    Source: openai.com

  3. Cursor New since July

    Anysphere, now part of SpaceX

    An AI code editor with agents, cloud agents and AI code review (Bugbot), now sold alongside xAI's Grok models.

    Pick it Choose it if you want to work inside an editor and steer agents line by line rather than hand off whole tasks.

    Price Free tier; Plus $8 a month (Grok models only); Premium $30 a month adds cloud agents and other makers' models billed at API price

    Source: cursor.com

  4. GitHub Copilot New since July

    GitHub (Microsoft)

    AI in the editor, the CLI and GitHub itself: agent mode, a cloud agent, code review and a Copilot app for multi-repo projects.

    Pick it Choose it when your team lives on GitHub and wants the cheapest paid entry point.

    Price Free tier; Pro $10, Pro+ $39, Max $100 a month, each with AI credits included

    Source: github.com

  5. Devin New since July

    Cognition

    An autonomous AI software engineer that works in the cloud, plus Devin Desktop (the next generation of Windsurf) as an IDE and command centre for agents.

    Pick it Choose it to hand a backlog of tickets to cloud agents that work through Slack, Linear or Jira, with an IDE for when you need to step in.

    Price Free tier; Pro $20 a month; Max $200 a month; Teams $80 a month plus $40 per developer seat

    Source: devin.ai

  6. Google Antigravity New since July

    Google

    Google's agent-first coding platform: a desktop app, CLI, IDE and extensions for other editors, running Gemini models.

    Pick it Choose it for a free starting tier, or when you already pay for Google AI Pro or Ultra and want Gemini 4 Argon.

    Price Free individual tier with weekly limits; higher limits with Google AI Pro or Ultra; business seats via Google Cloud

    Source: antigravity.google

What changed

  1. 28 Oct 2026 GitHub began including Copilot with GitHub Enterprise licences and added projects to the Copilot app, where several agents work across repositories. github.blog
  2. 13 Oct 2026 Gemini 4 Argon arrived in Google Antigravity for Google AI Pro and Ultra subscribers. antigravity.google
  3. 29 Sep 2026 At DevDay OpenAI launched Codex cloud environments, a refreshed Codex CLI, code review in the desktop app, Codex Security Cloud, and Sign in with ChatGPT so a ChatGPT plan pays for tools such as Devin, Amp and Conductor. openai.com
  4. 22 Sep 2026 Anthropic released Claude Opus 5.5, which it says leads in agentic coding and costs 40% less to run than Opus 5, and raised Claude Code usage limits on paid plans. anthropic.com
  5. 10 Sep 2026 Cognition released SWE-2, its own coding model, free to use on Devin's paid plans. cognition.ai
  6. 20 Aug 2026 Google Antigravity added extensions for other IDEs and became available to Gemini Enterprise customers. antigravity.google
  7. 14 Aug 2026 SpaceX completed its acquisition of Cursor, which now builds and ships xAI's Grok models inside the editor. cursor.com

Questions people ask about it

Vibe Coding

  1. Why can non-technical founders now build software with AI?
  2. What tools should I start with (Cursor, Claude Code, Lovable, Replit, Bolt, v0)?
  3. How do real founders ship an MVP with AI in days instead of months?
  4. What are the limits, when do I still need a real engineer?
  5. What actually counts as vibe coding, and how is it different from serious AI-assisted engineering?
  6. How do I plan and prompt so the AI builds what I actually want?

AI Agents

  1. What actually is an AI agent, in plain language?
  2. Why would a founder use agents instead of a chatbot or automation?
  3. What real tasks are founders delegating to agents today?
  4. How do I build my first agent without being technical?
  5. Which platform or framework should I build my agents on?
  6. How much do AI agents cost to run, and how do I keep the bill under control?

The rest of the map

Checked 29 October 2026. Prices and rankings move every few months; each tool links the page its claims were read from.

eChai Partner Brands The partnership →