we_are_coded.by CODE · The world, decoded
БГ
ElevenLabs

ElevenLabs shipped a command line written for agents, not people

ElevenLabs BlogAudio

ElevenLabs' new command line puts their whole product into the terminal: music, text to speech, agents. Responses come back as structured JSON, and errors are written so a program can parse them. Its user is another agent. The human comes second.

In short
  • The whole ElevenLabs API moves into the terminal: music, text to speech, listing agents, with output in structured JSON.
  • Agents as code: you pull the agent into local config files, edit it like code, and push it back to production with a review.
  • There's a dry-run mode that shows what would change without changing anything.
Checked on25 August 2026Responsible editorTsvetelin IvanovHow we workMethod · Corrections

Open the docs and the first thing you notice is what's missing. No confirmation prompt. No colored menu. Not one place where the tool waits for a human to press something.

The facts: ElevenLabs shipped version 1 of its command line. The whole product runs through it, that is making music, text to speech and listing agents, with output in structured JSON. Agents are pulled from the workspace into local config files, edited like any other piece of code, and pushed back to production with a review beforehand. The company calls it agents as code. The tool is designed agent-first: machine-readable schemas and errors in JSON that an agent can act on, instead of prompts written for a human. There's a dry-run mode that shows exactly what would change without changing anything, and it ships skills that teach the coding agent whole workflows. Source: the ElevenLabs blog.

Who it's written for

Their reasoning is short, and you can check it by glancing at your own screen: coding agents already live in the terminal. The tool just started speaking their language.

The user of this command line is another program. You come second.

The dry-run mode is the part I like most, and not because of the agents. A tool that can show you exactly what it will break before it breaks it is a well-built tool, no matter who's pressing the keys.

The visual is generated code art. No third-party images.
Follow usFacebookLinkedIn
Official primary sources
→ElevenLabs Blog: ElevenLabs CLI v1
Original: https://wearecoded.com/en/articles/elevenlabs-komanden-red-za-agenti.html
ShareFacebookXLinkedInTelegramWhatsApp

More on this

Audio
Security
Docker19 August 2026Coding agents

You approve a harmless command and somebody else's code runs. The hole is in the allowlist

Docker walks through a flaw in Cursor that let hidden text change environment settings without ever asking for approval. After that, an entirely ordinary command the developer approved themselves was running somebody else's code.

Read →
Frontier
SpaceXAI21 September 2026Models

Grok 4.7 is a bigger model at the old price - and it ships with a table that shows where it falls behind

On 21 September SpaceXAI released Grok 4.7: a new, larger base model, trained for longer, weighted toward tasks that take hours, at the same price as Grok 4.6. In the company's own table Fable 5.1 stays ahead on CursorBench and in the terminal, but it costs several times more.

Read →
Builders
Cloudflare3 August 2026Agents

Cloudflare gives the agent a computer: files, git and shell without a container for everyone

Agents Week opened with empty talk. @cloudflare/computer is the first real meat - an open library that gives the agent a file system, git, and code execution. The lightweight mode carries most of the work; a real Linux container only fires up for the heavy stuff.

Read →
← Back to all news