On October 16, an intruder reached a legacy system at OpenCode, potentially exposing third-party AI provider keys that users had saved in OpenCode Console. The hosted service stores keys for OpenAI, Anthropic, or Google so team members can use models without seeing the keys. OpenCode emailed affected users the next morning with the subject line "Action Required: Potential provider key breach."
The affected group is 1.4% of Console users, all with an outside provider linked. Keys stored in the command-line and desktop apps are local and not involved. Because provider keys are bearer credentials, anyone holding a copy can run billable requests against the owner's account. Deleting the key from Console does not revoke it, and a Console budget cannot cap usage that bypasses Console's gateway.
OpenCode advised revoking keys in the provider's dashboard, creating new ones, updating applications, and reviewing usage back to at least October 16. Some users report unauthorized charges, including hundreds to thousands of dollars. OpenCode has not confirmed any key was copied or used, and has not said when access began. The company described the incident as "related to a large-scale distillation attack" and said it is working with OpenAI, Anthropic, and Google.