OpenCode, the company behind an AI coding assistant, says an intruder reached a legacy system behind its optional hosted service, OpenCode Console, on the night of October 16. The system may have exposed third-party AI provider keys that users had saved in Console, where a workspace stores its own OpenAI, Anthropic or Google key so team members can use the models without ever seeing the key.
OpenCode says the affected group is 1.4% of Console users, all of whom had an outside provider linked, and that it emailed them on the morning of October 17 under the subject “Action Required: Potential provider key breach.” Anyone who did not get that email was not affected, according to the company. Co-founder Dax Raad said the breach involves “the OpenCode Console, not OpenCode itself,” and that keys added in the command-line and desktop apps “never come to us.” Those apps keep keys in a local SQLite database on your machine.
A provider key is a bearer credential: anyone holding a copy can run billable requests on the owner’s account until the provider revokes it. Deleting the key from Console does not revoke it. OpenCode’s own wording is that keys “may have been exposed”; it has not said any key was copied or used, nor how the legacy system was reached, how many keys it held, or when the unauthorized access began. It also called the incident ongoing and described it as “related to a large-scale distillation attack” without explaining the link. Some users have publicly reported unexpected charges on their own accounts, ranging from a few hundred to about $1,700, but those are individual claims, not confirmed by OpenCode.
How to check if you’re affected
Affected products are OpenCode Console workspaces with an OpenAI, Anthropic or Google key saved in them. The OpenCode CLI and desktop apps are described as not involved. If you got the “Potential provider key breach” email, act on it; OpenCode says no email means you were not affected. Steps from the company’s email and the providers’ guidance:
- Revoke the key in the provider’s own dashboard and create a new one. Deleting it in Console is not enough.
- Update every application and script that used the old key.
- Review usage and billing at the provider for requests you did not make, going back at least to October 16.
- Set a spend limit or alert at the provider and give each tool its own key, so a copied key does less damage and is easier to trace.
Sources
- OpenCode’s October 17 post on X
- Anthropic: what to do if you suspect your API key has been compromised
- Google Cloud: API key best practices
- Dealroom: OpenCode breach drains users’ AI keys (reports of user losses)
- The Hacker News: OpenCode Console Breach May Have Exposed OpenAI, Anthropic, and Google API Keys