docs(openrouter-images): sync with openrouter-web changes - #167
docs(openrouter-images): sync with openrouter-web changes#167devin-ai-integration[bot] wants to merge 1 commit into
Conversation
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
Original prompt from Automation
|
🤖 Devin AI EngineerI'll be helping with this pull request! Here's what you should know: ✅ I will automatically:
Note: I can only respond to comments from users who have write access to this repository. ⚙️ Control Options:
|
Summary
nonPOST /api/v1/imagesis now documented upstream as an upper bound, not a quantity: providers may return fewer images than requested, and for some models (e.g. Seedream 5.0 Lite) the returned count is prompt-dependent andndoes not influence it. The skill described--nas "Number of images to generate", which sets the wrong expectation for an agent that then reports or saves N images.Source: openrouter-web#33696 (
packages/image-generation/schemas/request.ts+ image-generation guide + OpenAPI:'Number of images to generate (1-10)…'→'Upper bound on the number of images to generate (1-10). Providers may return fewer images…').Changes in
skills/openrouter-images/SKILL.md:--nflag row restated as an upper bound; single-image providers still rejectn > 1.countreflects what the provider actually returned, and that multi-image results may need to be asked for in the prompt.No other public surface in the 24h window affects this skill: openrouter-web#33666 makes
output_formatdefault to the provider, which the skill already documents as "Model default".Link to Devin session: https://openrouter.devinenterprise.com/sessions/76b4ffec49104acdab4c6f2dc2619976