OpenCode says breach may have exposed linked provider API keys
OpenCode confirmed a security breach on October 17 and said third-party provider keys saved in some affected workspaces may have been exposed. The potentially exposed credentials include keys for OpenAI, Anthropic and Google.

Key takeaways · 4
- 01
The reported 1.4% refers to OpenCode Console users, with affected users drawn from those who linked external providers.
- 02
Keys saved in affected workspaces may have been exposed, including credentials for OpenAI, Anthropic and Google.
- 03
OpenCode asked affected users to revoke and replace provider keys, update applications that use them, and review recent usage.
- 04
OpenCode said it patched the issue in version 1.1.10 and notified affected users individually by email.
What OpenCode disclosed
OpenCode confirmed the breach on October 17 and described it as a large-scale distillation attack.[1][2] The company said the incident was ongoing and that it could not share more details at that point.[2] It also said it was working with Anthropic, OpenAI and Google to remediate the incident.[2] The disclosure establishes the reported scope and the potential exposure, but does not give further details about how the attack occurred. For teams assessing their response, distinguish the company’s description of the incident from details OpenCode said it could not yet share.[2]
Which users and keys are in scope
OpenCode said 1.4% of OpenCode Console users were affected.[2] OpenCode warned that third-party provider keys saved in affected workspaces may have been exposed; the named providers were OpenAI, Anthropic and Google.[2][1] The disclosure describes possible exposure, not confirmation that every listed key was accessed or misused. Users who did not receive an email from OpenCode were told they were not affected.[2]
Steps OpenCode recommended
OpenCode said it notified affected users individually by email and asked those users to revoke and replace their provider keys.[2] It also advised updating applications that use the keys and reviewing recent usage for unexpected activity.[2] Those steps matter beyond the OpenCode workspace: replacing a key alone may not update applications that were configured to use it, so affected teams should account for those applications when carrying out the company’s guidance. OpenCode said it was working with the named providers on remediation.[2]
Patch and user reports
OpenCode patched the issue in version 1.1.10.[2] Separately, users reported unauthorized charges of $400 on an Anthropic account and Google usage charges of around €200, with one user also saying their project was suspended for excessive abuse.[3] Some users said their OpenCode service API keys expired and social sign-in attempts failed after the breach.[3] These are user reports; the evidence does not establish that each reported charge or access problem resulted from the breach. OpenCode also said the incident was ongoing when it could not yet share more details.[2]
Teams using external provider credentials should treat this as a reminder to know where shared keys are stored and which applications depend on them. For affected users, following OpenCode’s key-rotation and usage-review guidance can help contain potential exposure without assuming that reported misuse applies to every account.
Why it matters
Put this to work — one session a day, built for your industry.
Create a free account for a daily session — eight questions and one real-work challenge, on the news that affects your role.
Start freeHow this developed
18 October 2026
OpenCode says breach may have exposed linked provider API keys
Sources
- OpenCode Breach: What Happened and How to Rotate Exposed AI API Keyscyera.com
- OpenCode Discloses Security Breach, Says Attackers May Have Accessed Users' OpenAI, Anthropic And Google API Keysofficechai.com
- OpenCode confirms breach as users report AI charges they never authorised - Notebookcheck Newsnotebookcheck.net