visual-summary-analysis
Performs AI analysis on input video clips/image content and generates a smooth, natural scene description. | 视觉摘要智述技能,对传入的视频片段/图片内容进行AI分析,生成一段通顺自然的场景描述内容
Preflight reference for HuggingFace snapshots — what vLLM, sglang, and transformers.generate see at runtime. Covers config-file precedence (tokenizer.json, tokenizer_config.json, generation_config.json, chat_template.jinja), transformers v5 tokenizer-class taxonomy (TokenizersBackend, PythonBackend, MistralCommonBackend, TikTokenTokenizer), special-token discovery (all_special_ids, added_tokens_decoder, extra_special_tokens, backend_tokenizer.get_added_tokens_decoder), chat-template Jinja contract (ImmutableSandboxedEnvironment, loopcontrols, raise_exception, strftime_now, tojson, add_generation_prompt), and engine knobs (skip_special_tokens, trust_request_chat_template, chat_template_kwargs allowlist, adjust_request, incremental detokenizer, EOS merge). Ships verified 2026 hall-of-shame for Kimi-K2.6, GLM-5.1, Gemma-4, Qwen3, DeepSeek-V3, plus drop-in Python for resolving markers to IDs, detecting turn-primer-as-EOS leaks, and cross-referencing tokenizer.json vs tokenizer_config.json.
This listing is imported from SkillsMP metadata and should be treated as untrusted until upstream source review is completed.
Install skill "transformers-config-tokenizers-expert" with this command: npx skills add air-gapped/skillsmp-air-gapped-air-gapped-transformers-config-tokenizers-expert
This source entry does not include full markdown content beyond metadata.
This detail page is rendered from real SKILL.md content. Trust labels are metadata-based hints, not a safety guarantee.
Related by shared tags or category signals.
Performs AI analysis on input video clips/image content and generates a smooth, natural scene description. | 视觉摘要智述技能,对传入的视频片段/图片内容进行AI分析,生成一段通顺自然的场景描述内容
全能高级前端研发工程师技能。擅长AI时代前沿技术栈(React最新 + shadcn/ui + Tailwind CSS v4 + TypeScript + Next.js),精通动效库与交互特效开发。采用Glue Code风格快速实现代码,强调高质量产品体验与高度友好的UI视觉规范。在组件调用、交互特效、全局Theme上保持高度规范:绝不重复造轮子,相同逻辑出现两次即封装为组件。具备安全意识,防范各类注入攻击。开发页面具有高度自适应能力,响应式设计贯穿始终。当用户无特殊技术栈要求时,默认采用主流前沿技术栈。
摩天轮开放接口二次封装技能。当用户需要对接摩天轮(MTL)开放平台接口、创建MTL HTTP接口、封装摩天轮API时触发。支持自动检测并添加mtl-open-sdk-consumer依赖、初始化ApiClient Bean、创建MtlRestController、根据原生接口定义生成HTTP接口代码。
Non-contact detection of heart rate, respiration, blood oxygen, and heart rate variability. No wearable devices are required; monitoring is achieved solely through camera footage. | 无感生命体征监测分析技能,非接触检测心率、呼吸、血氧、心率变异性,无需穿戴设备,通过摄像头画面即可监测