ElevenLabs rebuilt its examples repo so changelog merges trigger agents that update skills, run evaluations, and refresh examples before human verification. The workflow turns developer example maintenance into an explicit multi-agent process instead of a manual docs task.
ElevenLabs is framing Examples 2.0 as infrastructure for keeping developer samples in sync with a fast-moving API. In the company's workflow post, a merged changelog becomes the handoff point: changes are summarized into an issue, an autonomous agent updates the relevant skills, and that pass is checked against the API reference before a human reviews and merges the PR.
Once that merge lands, the pipeline branches again. ElevenLabs' evaluation post says one agent updates the examples repository while another runs a skills evaluation suite, and the company adds that the eval suite was built with Claude's "skill-creator skill," suggesting the maintenance stack is using generated tooling to validate generated updates.
The last stage is a repo-wide sweep. ElevenLabs says in its automation post that "a script loops through all examples and updates them if needed," with the whole flow currently running through Cursor Automations in the cloud. The published repo at GitHub and ElevenLabs' minimal design post also show the output target: minimal examples that "showcase only what is necessary" for each use case, rather than broader tutorial-style projects.
Introducing ElevenLabs Examples 2.0, a modern approach to the examples repository. The examples repo is one of the highest touch points for any developer-facing product, but maintaining them is a tedious task when you have a fast moving product. We rebuilt ours to be Show more
The last step is an autonomous agent updating the examples. A script loops through all examples and updates them if needed. All of these currently run in the cloud using Cursor Automations.
