feat(yollomi): add new commands and update documentation in README files#235
feat(yollomi): add new commands and update documentation in README files#235jackwener merged 3 commits intojackwener:mainfrom
Conversation
- Added yollomi commands for generating images, videos, and editing capabilities. - Updated README.md and README.zh-CN.md to include yollomi in the command list. - Enhanced SKILL.md with yollomi-related tags and usage examples.
- Included yollomi in the VitePress configuration for browser adapters. - Updated adapters index documentation to reflect yollomi's capabilities and commands.
Astro-Han
left a comment
There was a problem hiding this comment.
Nice work covering so many commands with a clean, consistent structure! A few things I noticed:
upload.ts — base64 in page.evaluate may crash on large files
The entire file is read into memory, base64-encoded, and embedded as a JS string literal inside page.evaluate(). For the 100MB video limit, that's ~133MB of base64 injected into the browser JS engine — likely to OOM or timeout. Consider lowering the size cap significantly, or noting this limitation in the docs.
models.ts — consider adding browser: false
This command returns hardcoded data and never uses page. Without browser: false, the framework may still attempt a browser connection, which would be unnecessary overhead for users just listing models.
edit.ts — unused import
resolveImageInput is imported but never called.
Tests
No tests are included. At minimum, models (PUBLIC strategy) could have a straightforward test in public-commands.test.ts, and the browser commands could have graceful-failure entries in browser-auth.test.ts per TESTING.md.
- models.ts: add browser: false (no browser connection needed for hardcoded data) - edit.ts: remove unused resolveImageInput import - upload.ts: lower video upload limit from 100MB to 20MB (base64 OOM risk) - generate.ts: improve file extension detection using URL.pathname - upscale.ts: use choices for scale arg, improve extension detection - object-remover.ts: make image/mask args positional - Add yollomi models tests to public-commands.test.ts - Add yollomi generate/video graceful-failure tests to browser-auth.test.ts
|
感谢 @AniChikage 贡献 yollomi adapter！覆盖了图片生成、视频生成、编辑等多种 AI 模型，结构非常清晰 🎉 在合并前我做了以下修复和改进： Bug 修复：
改进：
测试：
再次感谢你的贡献！ |
Description
Related issue:
Type of Change
Checklist
Documentation (if adding/modifying an adapter)
docs/adapters/(if new adapter)docs/adapters/index.mdtable (if new adapter)docs/.vitepress/config.mts(if new adapter)Screenshots / Output