tools

OpenCode says a breach exposed linked AI provider keys for 1.4% of its Console users

The open-source coding agent's hosted Console was hit by what OpenCode calls a large-scale distillation attack. Affected users were emailed and told to replace their OpenAI, Anthropic and Google keys.

Written by Mara at Sig · Oct 19, 2026

OpenCode said on X on October 17 that it had a security breach the night before, "related to a large scale distillation attack that impacted 1.4% of OpenCode Console users." It said everyone affected was notified by email that morning, that users who did not get an email were not affected, and that only a subset of users who had linked external providers were hit. It said it was working with Anthropic, OpenAI and Google to remediate.

Security firm Cyera, summarizing the incident, says OpenCode's email to affected users tied the intrusion to a legacy console system the company was in the process of retiring. OpenCode has not said how attackers got in or how the keys were stored, and Cyera reports it can't yet confirm whether exposed keys were misused. Some users report usage charges they didn't make.

OpenCode Console is the hosted layer where teams share models from a workspace, set budgets and connect their own provider keys. Cyera points out that Console budgets only cover spend that runs through OpenCode, so they don't limit a stolen key used directly against the provider.

Why it matters

Any hosted tool that stores your provider keys is now a target, because a stolen key means free access to expensive models, which is exactly what someone distilling a model needs. Rotate keys at the provider itself and set hard spend limits there.