Anomalyco/OpenCode requires x-opencode-session through LiteLLM in just 3 days

[nmapx]

OpenCode just asked LiteLLM to send x-opencode-session header on all out outbound inference HTTP requests to their API.
BerriAI/litellm#39503

Since there is no "opencode" provider in LiteLLM (which means there is none for them in OpenHands too) I'm not sure how they are going to do that, if at all. IMO LiteLLM should either recognize OpenCode API itself or accept a header (eg. from OpenHands) to pass it further to the inference provider API. Both solutions will require OpenHands release due to possible LiteLLM update which is a bit tricky since we are short on time. Opencode announced it today morning with a deadline on 6/09/2026. Just crazy.

I'm opening this issue to track this topic. All comments are more than welcome.
Cheers!

[github-actions]

⚠️ Not yet ready-for-dev

This issue does not yet meet the readiness criteria, so the
ready-for-dev label has not been applied (or has been removed).
The criteria are type-specific:

Bug reports (bug label, see .github/ISSUE_TEMPLATE/bug_template.yml):

• The ### Actual Behavior section must describe reproducible SDK steps
  and include a command such as python, pytest, uv, or pip.
• An ### Acceptance Criteria section with at least one checklist item (- [ ] …).

Enhancements (enhancement label, see .github/ISSUE_TEMPLATE/feature_request.yml):

• A ### Desired Behavior section.
• An ### Acceptance Criteria section with at least one checklist item.

What is missing:

• The issue has neither the bug nor enhancement label, so its readiness criteria cannot be evaluated. Add the appropriate label.

Edit the issue to address the gaps and the ready-for-dev label will be
applied automatically.

Note: Section headings require exact ### matches from the templates.
If a missing type label is reported, no action is needed from external
contributors — maintainers will add the label during triage.

This is an automated check — no AI was used to generate this comment.

[neubig]

Thanks, we're looking at this.

[rbren]

@nmapx are you using OpenCode LLMs with the OpenHands harness?

[nmapx]

@rbren yes

[nmapx]

FYI: OpenCode is already enforcing new API validation.
No solution on LiteLLM as far as I know, yet.