ElevenLabs CLI 1.0: scripting audio with Sume over plain HTTP

ElevenLabs CLI v1.0.0 exposes every ElevenLabs API operation as a subcommand with JSON, table, YAML and CSV output. This post is about the Sume side of that shell workflow: audio is plain HTTP, with an Authorization: Bearer header, an Idempotency-Key, and job URLs you poll.

CLI facts are from the ElevenLabs changelog for August 24, 2026. Sume facts are from Timeline audio and Jobs and results, read 2026-10-01.

What does the ElevenLabs CLI 1.0 cover?

Per the changelog: every API operation as a subcommand, automatic pagination and shell completion, plus file-based configuration of agents that you push and pull. It adds nothing you need to install to call Sume; Sume's documented surface is HTTP.

How do I run a Sume audio job from a shell?

Join two Sume-hosted clips with timeline audio, then poll the job. Parts must already be this workspace's media.sume.com audio, so import first with POST /v1/media-imports. The response gives a job id; GET /v1/timeline-1.0/audio/:id does not exist, so poll the job envelope.

curl -X POST https://api.sume.com/v1/timeline-1.0/audio \
  -H "Authorization: Bearer $SUME_API_KEY" \
  -H "Content-Type: application/json" \
  -H "Idempotency-Key: audio-join-001" \
  -d '{"operation":"concat","parts":[
    {"url":"https://media.sume.com/artifacts/artf_demo/line1.wav"},
    {"url":"https://media.sume.com/artifacts/artf_demo/line2.wav"}]}'

curl https://api.sume.com/v1/jobs/job_123/result \
  -H "Authorization: Bearer $SUME_API_KEY"

What are the limits and the price?

Which output should I keep for scripts?

Keep wav when the file will be joined again, since mp3 re-adds priming padding at every edge. The docs say to confirm the live rate in GET /v1/catalog. If you also drive agents from a terminal, the same jobs are reachable through hosted MCP.