mirror of
https://github.com/lobehub/lobehub.git
synced 2026-03-26 13:19:34 +07:00
e67bcb2571
* ✨ feat(cli): add generate command for text/image/video/tts/asr LOBE-5711 - `lh generate text <prompt>` — LLM text completion with SSE streaming - Supports --model (provider/model format), --system, --temperature, --pipe - `lh generate image <prompt>` — Image generation via async task - `lh generate video <prompt>` — Video generation via async task - `lh generate tts <text>` — Text-to-speech (openai/microsoft/edge backends) - `lh generate asr <file>` — Speech-to-text via OpenAI Whisper - `lh generate status` — Check async generation task status - `lh generate list` — List generation topics - Add shared HTTP auth helper (api/http.ts) for webapi endpoints Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com> * update info * ♻️ refactor(cli): split generate command into submodules, text defaults non-streaming - Split monolithic generate.ts into generate/{index,text,image,video,tts,asr}.ts - Text subcommand now defaults to non-streaming (use --stream to opt in) - Text subcommand supports --json for full JSON response output - Video subcommand uses requiredOption for --model and --provider Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com> * 🐛 fix(cli): read generation data from result.data and add required X-lobe-chat-auth header Image/video mutations return { success, data: { ... } }, read IDs from data. WebAPI endpoints require X-lobe-chat-auth (XOR-encrypted) alongside Oidc-Auth. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com> --------- Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>