Guides · Claude Code and Codex for teams
Claude Code guides for teams
These guides answer what a team asks before it runs Claude Code and Codex for more than one person: what it costs, what can stop a session, how to see usage per person, where the keys and the data go, and how to run other models. The comparisons put Selan beside the tools a team weighs it against, and each says where the other tool is the better answer.
What does Claude Code cost a team?
- How much does Claude Code cost per developer?: Anthropic's own figures a day and a month, why developers differ, and how to measure and cap it.
- Is a Claude Max plan cheaper than the API?: where a seat costs less than paying per token.
- How does prompt caching change what Claude Code costs?: a cache write against a read, model by model, and what makes Claude Code miss.
How do you see and limit usage?
- What do you do when Claude Code says rate limit reached?: the 5-hour and weekly windows, when each resets, and the ways past them.
- What do you do when Claude Code says spend limit reached?: whose limit each message names, what lifts it, and a limit per developer.
- How do you see Claude Code usage per developer?: per person and per team, with Codex and Claude Desktop in the same figures.
- How do you track Codex CLI usage?:
/status, OpenAI's workspace analytics, and Codex beside Claude Code per person.
How do you run it on company accounts?
- How do you keep API keys off developers' laptops?: keys on a gateway, and a sign-in per person instead.
- How do you give a team Claude Code and Codex with EU data residency?: which providers run in the EU, and what can still leave.
- How do you run Claude Code in GitHub Actions for a team?: which credential to store, how to cap a run, and a token per job.
- How do you skip permissions in Claude Code safely?: what the flag turns off, where to run it, and the limits a team keeps around it.
- What should an AI coding tool policy say?: ten clauses for Claude Code and Codex, each matched to the setting that enforces it.
How do you run other models in Claude Code?
- How do you run GPT in Claude Code?: why it needs a gateway that translates, and the routes to OpenAI.
- How do you run Kimi K3 in Claude Code?: Moonshot's own setup, then a team's on OpenRouter or Ollama Cloud.
- How do you run DeepSeek in Claude Code?: DeepSeek's own setup, then a team's on Ollama Cloud or OpenRouter.
- How do you use Ollama with Claude Code?: local or cloud by Ollama's setup, and Ollama's cloud for a team.
How does Selan compare?
- Selan and ccusage: one machine's logs against a team's usage per person.
- Selan and LiteLLM: a proxy you run for your own apps against a hosted gateway for coding tools.
- Selan and Anthropic's Claude apps gateway: self-hosted on your own cloud against hosted with Codex and subscriptions.
- Selan and OpenRouter: models per token against your subscriptions first, with OpenRouter behind them.
- Selan and nexos.ai: a company AI platform in credits against your subscriptions, with your nexos.ai key as a route.