skills

阿里云百炼平台技术文档知识库（LLM Wiki）。当用户查询百炼模型列表、API 参数、错误码、 应用开发（智能体/RAG/知识库/记忆/插件）、模型对比与定价、SDK/OpenAI 兼容接口、 多模态能力（语音/图片/视频）、Token 计费等百炼相关问题时激活。 包含 models 结构化模型市场数据（含 contextWindow/QPM/价格/sample code）、 wiki 合成层（主题页/概念页/对比页）和 raw 原始文档层； 模型规格类问题先查 models/index.md，文档类问题先查 wiki/index.md。

阿里云百炼模型选型与推荐。当用户需要"选择、推荐、对比"模型，或描述一个 AI 场景/功能需求（隐含需要帮其决策用哪个模型）时激活。核心意图是帮用户做决策， 不是单纯查资料。触发词：推荐模型、选哪个、哪个适合、对比、做一个XX、实现XX 功能、用什么模型好、XX场景方案。当用户同时涉及模型查询和模型选择时，优先 使用本 skill（本 skill 内部会读取模型数据完成推荐）。

用百炼 CLI (`bl`) 走完"数据→微调训练→导出→部署→调用"的完整闭环，或跳过训练直接部署基座模型。支持文本模型（SFT/DPO/CPT）、音频 TTS 模型（CosyVoice）、图像生成模型（Wan2.7）和视频生成模型（Wan i2v/kf2v）微调。涵盖数据集校验/上传、创建微调任务、等待训练、导出最佳 checkpoint、创建推理部署、等待就绪、给出调用示例。当用户提到在百炼 / DashScope / 阿里云模型工作室上"训练模型""微调""fine-tune""finetune""部署模型""模型上线""把微调模型跑起来/调用""训练一个推理模型""继续预训练""LoRA/SFT/DPO 训练""语音合成模型""TTS 微调""CosyVoice""声音克隆""图像生成微调""文生图""图生图""Wan2.7""图像模型训练""视频生成微调""图生视频""首帧生视频""首尾帧""kf2v""i2v""Wan视频""视频特效训练"等，都应激活本技能——即使用户没明说"用 bl"，只要意图是百炼平台的训练或部署，就用本技能，不要自己拼凑命令。

Use when users need China or Hong Kong securities, funds, fund managers, company financials, valuation, global macro or industry time series, broker research, announcements, financial news, or enterprise-risk data. Triggers include 选股、基金筛选、基金经理、净利润、营收、ROE、估值、 GDP、CPI、核心 PCE、行业产销价、研报、公告、财经新闻、工商与司法风险。

Interactive prompt studio for HappyHorse 1.0 video generation. Guides users through scenario discovery with vivid examples, then assembles production-ready prompts in JP/CN/EN. Use when someone wants to create AI video content with HappyHorse but doesn't know where to start, or when they have a specific scenario and need a polished prompt. Covers manga drama, character PV, manga motion, virtual idol MV, and free-form scenarios.

将小说/故事改编为互动小说网页游戏（React SPA），含 AI 生成素材（视频/图片可选）、 可选 TTS 旁白、程序化音频、分支剧情引擎、存档系统。 使用 `bl` CLI 完成素材生成（`bl video generate` / `bl image generate` / `bl speech synthesize`）。 当用户提到互动小说、文字冒险游戏、小说改编游戏、H5 互动游戏、分支剧情、视觉小说时激活。

One-shot autopilot orchestrator — runs the full spark-video pipeline (screenwriter ↔ director per-scene parallel → render chain-DAG parallel + per-clip review → stitch). User confirms at 4 gates (+ 1 mode gate at start + 1 BGM gate when bgm/ folder detected). Use when the user wants "make me an episode" in one command.

Turn ONE topic into a finished Vox-style paper-collage explainer / ad video, end to end with Aliyun Bailian CLI + local ffmpeg — script, collage keyframes, motion, voice-over, music, captions, all automated. Use this whenever the user wants a "Vox style" video, a paper/torn-paper collage animation, a "motion collage", a narrated explainer or short ad built from AI-generated collage posters, a scrapbook-style tribute, or wants to turn a topic / product / person into a punchy narrated collage video — even if they don't say the word "Vox". Also use when reproducing Stav Zilber / rom1trs / Higgsfield-style collage ad workflows. Three input modalities: a topic (B-roll), a talking-head video (A-roll mode), or a single photo of a person/product anchored into the collage (C-roll mode). Triggers: "vox video", "collage video", "motion collage", "paper collage explainer", "make a collage ad", "turn this topic into a collage video", "turn my photo/this product shot into a collage video".