[waleedlatif1]

Summary

• Preserve Agent-supplied fal video settings while keeping configured overrides and explicit model defaults authoritative.
• Use synchronous Google Omni URI delivery and wait for Files readiness before the bounded download, avoiding oversized inline polling responses.
• Reject MiniMax H3 legacy quality options before paid submission and describe legacy-only prompt optimization accurately.
• Document the current Image Generator's existing 4,000-character prompt ceiling separately from provider limits.

Follow-up to the media changes in #9640, #9653, and #9658, including the relevant review findings on #9660.

Type of Change

• Bug fix

Testing

Regression cases were shown failing before their fixes at the actual Agent/tool/provider-request and provider-response boundaries. Compatibility cases cover configured overrides, explicit defaults, false booleans, legacy models, file processing failure, cancellation, URI validation, and bounded metadata.

Passed root lint, all 29 workspace type checks, the unchanged test-type ratchet, all 62 audits (after generating the fresh checkout’s docs registry), block-registry validation, and the full root test command: 727 script tests, 39,484 app tests, and all 20 workspace tasks. The 291 focused media/serializer/block tests also passed. Initial unrelated script-test timeouts passed on the lower-concurrency rerun without code or timeout changes. No paid provider inference requests were submitted.

Checklist

• Code follows project style guidelines
• Self-reviewed my changes
• Tests added/updated and passing (new tests pass the test-audit authoring gate)
• No new warnings introduced
• I confirm that I have read and agree to the terms outlined in the Contributor License Agreement (CLA)

[vercel]

The latest updates on your projects. Learn more about Vercel for GitHub.

1 Skipped Deployment

Project	Deployment	Actions	Updated
docs	Skipped		Nov 4, 2026 10:19am UTC

[waleedlatif1]

@greptile

[waleedlatif1]

@cubic-dev-ai review this PR

[cubic-dev-ai]

@cubic-dev-ai review this PR

@waleedlatif1 I have started the AI code review. It will take a few minutes to complete.

[greptile-apps]

Confidence Score: 5/5

[Medium impact] The PR appears safe to merge; no actionable new issues were found.

Summary

This PR preserves Agent-supplied video settings, updates Google video delivery, and clarifies model limits.

• Image-generation prompts show Sim’s 4,000-character limit.
• Fal video requests keep Agent settings unless the block overrides them.
• Gemini video generation returns a file link that Sim downloads when ready.
• MiniMax H3 uses resolution settings instead of legacy quality options.

_{Reviews (2) · Last reviewed commit: "fix(image): keep legacy prompt documenta..." · Reviewed by Greptile}

[cubic-dev-ai]

All reported issues were addressed across 10 files

<sub>Reply with feedback, questions, or to request a fix.

Turn on auto-fix | Re-trigger cubic</sub>

[waleedlatif1]

@greptile

[waleedlatif1]

@cubic-dev-ai review this PR

[cubic-dev-ai]

@cubic-dev-ai review this PR

@waleedlatif1 I have started the AI code review. It will take a few minutes to complete.

[cubic-dev-ai]

No issues found across 10 files

Confidence score: 5/5

• Automated review surfaced no issues in the provided summaries.
• No files require special attention.

_{Turn on auto-fix | Re-trigger cubic}