Codex + Azion
Terminal agent that runs commands in a sandbox and reads AGENTS.md natively. Azion goes in config.toml. Made by OpenAI.
Last updated
View as MarkdownQuick start
Install the CLI globally and sign in with your OpenAI account.
The authenticate as you. Create a in the Console, give it only the scopes you want Codex to have, and copy it — it is shown once.
Codex reads one TOML configuration for every project. Add the server there and it is available in each sandbox it opens.
Start a session and ask which tools it has. The seven docs server tools, such as search_azion_docs_and_site, are listed alongside its own shell and file tools.
Try it
Azion platform access
Three layers, and a connected agent uses all three: the MCP servers to look things up and create configuration, the CLI to build and deploy from your working copy, and a context file so it starts each session knowing your account.
MCP servers
Five servers, one per area of the platform. Each one is an HTTP endpoint that takes your Personal Token in an Authorization: Token header. The quick start connects the docs server. Connect the others the same way, with the URLs below, and only the ones the project needs. The docs server only reads. The other four also create, change, and delete, and their write tools preview a change without running it when the agent sets dry_run:
Azion CLI
Local builds, deploys and the product commands the API does not cover. Install it once and Codex runs it for you — the first azion login is interactive, so run that one yourself.
Agent context
Codex reads AGENTS.md at the repository root before it answers. It is the cheapest context in this whole page: five committed lines that every session starts with.
Agent-friendly docs
Token-efficient references an agent can fetch on its own, with no tool call and no token. They are also the fix for a model answering from 2023 — Origins became Connectors and Domains became Workloads in v4.
No MCP support in the tool you are using? Prime it by hand, once per session:
Any assistant
Example prompts
Each of these exercises a different half of the connection — the documentation search, the CLI, the configuration it writes, the analytics API, the storage.
Tips
FAQ
Both, and they do different jobs. The are how Codex looks things up and creates configuration through the API; the is how it builds and deploys from your working copy. Ask for a deploy and a connected agent will reach for the CLI on its own.
Troubleshooting
Check the header before anything else — a prefix other than Token or an expired token fails as a silent 401 or 403, not as an error in the tool. Then restart the client: every one of them reads its MCP configuration at startup.