NNNSX

Generate images with Aihubmix APIs, especially openai-compatible gpt-image-2 and GPT-Image-1 image generation. Supports prompt files, OpenAI-compatible image generation, reference-image editing in openai mode, optional zimage native mode, output downloads, dry-run payload inspection, and saved JSON responses.

Generate images with Shengsuanyun Router using openai/gpt-image-2, including prompt-file generation, local or URL reference images, task polling, downloads, and saved JSON responses. Use when the user wants Codex to call the Shengsuanyun image API, use gpt-image-2 through router.shengsuanyun.com, or package/reuse this API-based image generation workflow.

Assist with research-paper writing and revision under local writing rules. Use when Codex is asked to draft, rewrite, polish, compress, expand, restructure, respond to reviewer comments, revise from an audit report, align with journal/conference requirements, write captions, improve experiment analysis, reduce overclaims, remove AI-like wording, or synchronize a manuscript, response letter, supplement, and final submission.

Review research-paper PDFs with multi-channel reading: extracted text, page/figure visual inspection, formulas, tables, data-reuse consistency, writing quality, and compliance with local thesis/paper writing rules. Use when Codex is asked to read,审查, audit, review, revise, or diagnose a PDF paper, especially when figures, formulas, tables, captions, experimental claims, or data provenance must be checked.

用轻量主会话作为 chief、一次性子会话作为 worker 来协调复杂任务。适用于重上下文、多模态文件、大文件树、资料检索、并行调查、批处理、provider 调用、文件处理，或任何需要主会话只保留决策、路径、链接、简短观察和最终汇报，而把有界执行交给 worker 完成并停止的工作流。

Create, package, submit, query, download, and review AI video-generation jobs through the official Xiaoyunque/Pippit CLI and Seedance models. Use when Codex needs to install or verify @pippit-dev/cli, use pippit-tool-cli generate-video/query-result, prepare Xiaoyunque video prompts, choose Seedance 2.0/1.5 model variants, work with text-to-video or image/video-reference generation, package reproducible job files, inspect generated videos, or document Xiaoyunque video generation attempts safely.

Generate and edit images through a sub2api-backed OpenAI-compatible GPT Image provider. Supports Responses API streaming image generation, GPT Image text-to-image generation, image editing with one or more reference images, optional alpha masks, JSONL batch generation, prompt files, quality/size/output-format controls, dry runs, provider diagnostics, model listing, dependency bootstrap, run logs, and project output paths.