The Zoltbot blog
Guides, comparisons, migration notes and short answers about running coding agents, from the team building Zoltbot. Every claim about another tool links to its source.
Guides to running coding agents
Longer pieces on running several coding agents, what they cost, and how to keep the work organized. We use these tools every day and write down what we learn.
How to run several coding agents at once
Give each agent its own git worktree and branch so their edits land in separate folders. Then plan for what a worktree leaves shared: dev server ports, the local database, chat history and your review queue. On one laptop that works for a few agents at a time, and past that the bookkeeping is the hard part.
ReadWhat coding agents cost: Claude Code, Codex, Cursor and API usage
The entry plans that include Claude Code (Claude Pro) and Codex (ChatGPT Plus) cost $20 a month, and Cursor's first plan with cloud agents and other vendors' models costs $30. Heavy users move to $100 or $200 plans. An API key has no monthly cap: Anthropic reports that its enterprise Claude Code users average about $13 per developer per active day.
ReadWhat is a coding agent workspace?
A coding agent workspace is an app that sits above your coding agents and keeps track of the work: which project a task belongs to, which agent runs it, where its conversation lives and whether its result has been reviewed. The agents keep doing the coding. The workspace gives them, and you, one place to organize it.
ReadLoop engineering: let loops prompt your coding agents
Loop engineering means you design the system that prompts your coding agents instead of prompting them yourself, turn by turn. A loop that can run without you needs six parts: a trigger, a written definition of done, a worker in its own worktree, a separate checker, a record of progress kept outside the conversation, and a way to stop or hand the work back to you.
Read
Coding agent comparisons
Claude Code, Codex and Cursor side by side: where each one does better, with prices checked against the official pages and every claim linked to its source.
Claude Code vs Codex
Choose Claude Code if you work in the terminal and want Claude's own tools for running many sessions at once. Choose Codex if you already pay for ChatGPT and like a desktop app that mixes local worktree chats with cloud tasks. Both start at $20 a month, and we use both every day.
ReadCursor vs Claude Code
Choose Cursor if you want an editor with agents built in, cloud agents in their own VMs, and models from several vendors. Choose Claude Code if you prefer the terminal and want Claude models on a flat monthly plan. Claude Code starts with Claude Pro at $20 a month; Cursor's $8 plan runs Grok models only, and cloud agents start at $30.
Read
Coding agent app alternatives, compared with Zoltbot
Leaving a coding agent app? See how Zoltbot compares, when to stay with the tool you have, and how to move your work, with every claim sourced.
Antigravity alternatives, or Antigravity beside Claude Code and Codex
The alternatives people weigh against Google Antigravity are Claude Code, Codex and Cursor. You can also keep Antigravity and use them beside it. Zoltbot is a free, open-source Mac app that runs Antigravity, Claude Code, Codex and Cursor through their own CLIs, and each task can get its own Git worktree. Stay with Antigravity alone if you work on Windows or Linux, or if its own App already runs all the agents you need.
ReadConductor alternatives: how Zoltbot compares
Conductor is joining Meta and shutting down. Conductor Cloud becomes read-only on 13 November 2026 and shuts down on 20 November. The Mac app gets fixes until early 2027, then shuts down and deletes all user data. Zoltbot is a free, open-source Mac app that runs Claude Code, Codex, Cursor and Antigravity side by side, each task in its own Git worktree. If you depend on Conductor's pull request and checks flow, finish current work there first.
ReadParadev (formerly Superset) alternatives: how Zoltbot compares
Paradev is the new name of Superset, the open-source workspace for running coding agents in parallel (not Apache Superset). If you're looking for a Superset or Paradev alternative on a Mac, Zoltbot is free and open source, runs Claude Code, Codex, Cursor and Antigravity, and gives each task its own Git worktree inside nested projects. Stay with Paradev if you need remote hosts, cloud workspaces, its phone app, automations or many more agents.
Read
Short answers about coding agents
Quick answers to questions we hear from people setting up Claude Code, Codex, Cursor and Zoltbot. Each one starts with the answer.
Can I use my Claude subscription with multiple coding agents?
Yes for Claude Code: every Claude Code session you start, in as many terminals or worktrees as you like, runs on the same Claude Pro, Max or Ultra plan and shares its usage limits. Other vendors' agents need their own plan, so Codex uses a ChatGPT plan or OpenAI API key, and Cursor bills Claude models at API list price outside its plans.
ReadCan I use my ChatGPT plan with the Codex CLI?
Yes. Run codex, choose Sign in with ChatGPT, and the CLI uses your plan's Codex allowance; the CLI is included from ChatGPT Plus at $20 a month. You can sign in with an OpenAI API key instead, which bills at API rates and leaves out Codex's cloud features.
ReadCan Claude Code and Codex work on the same repo at the same time?
Yes, as long as they don't share a checkout: give each agent its own git worktree and branch, and merge one branch at a time. Without worktrees, run them one after the other. Zoltbot can do either: it gives a task its own worktree, or queues tasks that share a folder.
ReadWill I lose my chats if I stop using Zoltbot?
No: each tool's own session store holds the conversation, and Zoltbot keeps only an index that points to it. If you stop using Zoltbot, the conversations are still where Claude Code, Codex, Cursor and Antigravity keep them.
Read