awesomewebpagemetaskill

---

name: AwesomeWebpageMetaSkill description: "Build a local multimedia webpage project from a user topic, audience, language, style, and media preferences by framing requirements, researching, planning, acquiring media, generating files, packaging, validating, repairing, and delivering usage guidance." description_zh: "根据用户给定的主题、受众、语言、风格和媒体偏好，通过梳理需求、调研、规划、获取媒体、生成文件、打包、校验、修复并交付使用指引，构建一个本地多媒体网页项目。" kind: meta meta_priority: 65 always: false final_text_mode: "step:delivery_guide" request_template: outcome: "A packaged local multimedia webpage project with assets, validation notes, and usage guidance." outcome_zh: "打包好的本地多媒体网页项目，包含素材、验证说明和使用指南。" outcome_en: "A packaged local multimedia webpage project with assets, validation notes, and usage guidance." fields: - name: topic label: "Topic" label_zh: "网页主题" label_en: "Topic" - name: target_audience label: "Target audience" label_zh: "目标受众" label_en: "Target audience" - name: output_language label: "Output language" label_zh: "输出语言" label_en: "Output language" - name: visual_style label: "Visual style" label_zh: "视觉风格" label_en: "Visual style" - name: media_preferences label: "Media preferences" label_zh: "媒体偏好" label_en: "Media preferences" triggers:

• "create awesome webpage"
• "build multimedia webpage project"
• "生成图文音视频网页"
• "做一个多媒体网页项目"
• "AwesomeWebpageMetaSkill" provenance: origin: opensquilla-original license: Apache-2.0 maintained_by: OpenSquilla metadata: opensquilla: risk: high capabilities: [network-read, network-write, filesystem-read, filesystem-write, command-exec] requires: config: - awesome_webpage.openrouter.models.page_generation - awesome_webpage.openrouter.models.image_generation - awesome_webpage.openrouter.models.audio_generation - awesome_webpage.openrouter.models.video_generation - awesome_webpage.output_dir - awesome_webpage.media_strategy config: awesome_webpage: openrouter: models: page_generation: moonshotai/kimi-k2.6 image_generation: google/gemini-3-pro-image-preview audio_generation: openai/gpt-audio-mini video_generation: bytedance/seedance-2.0-fast output_dir: "{{ inputs.workspace_dir }}/awesome-webpage-output" media_strategy: search_first: true aigc_fallback_when_search_empty: true allow_remote_embeds: false default_modalities: [text, images, audio, video] search_modalities: [images] direct_aigc_modalities: [audio, video] confirmation_steps: [ask_images, ask_audio, ask_video, ask_style, media_provider_approval] aigc_policy: search_images_direct_generate_audio_video placeholder_policy: visible_replacement_slot_when_generation_unavailable target_assets: images: 6 audio: 1 video: 1 licensing: prefer_reusable_or_user_supplied clawhub_skills: web_search: skill: web-search url: https://clawhub.ai/billyutw/web-search image_generation: skill: nano-banana-pro-openrouter url: https://clawhub.ai/skills/nano-banana-pro-openrouter opensquilla_compatibility: deterministic-skill-exec notes: | The image_aigc step runs this deterministic skill_exec adapter. It does not invoke a prompt-building skill as an agent. Its volatile connection comes from ordinary Provider Settings after approval. audio_generation: skill: audio-cog url: https://clawhub.ai/skills/audio-cog opensquilla_compatibility: openrouter-config-first video_generation: skill: openrouter-video-generator url: openrouter-configured-video-generation webpage_generation: skill: html-coder url: https://clawhub.ai/jhauga/html-coder opensquilla_compatibility: scoped-agent notes: | Invoked by the webpage_generation step as the HTML/CSS/JS authoring skill. The meta-skill task constrains it to source JSON only; packaging, validation, repair, and media generation remain separate steps. filesystem: skill: filesystem url: https://clawhub.ai/gtrusler/clawdbot-filesystem composition: steps:

  • id: requirement_framing kind: llm_chat with: system: | You are the Requirement Framing stage for AwesomeWebpageMetaSkill. Produce a compact, structured brief. Do not call tools. task: | {{ inputs.language_instruction }}

    Extract and normalize the user's webpage request.
    
    Required fields:
    - topic
    - target_audience
    - output_language
    - visual_style
    - media_preferences: infer the user's requested images, audio, video, remote embeds, and local-only needs
    - configured_output_dir: use the resolved config below; only mark CONFIG_NEEDED if it is empty
    - constraints and risks
    
    Do not invent provider names, model ids, API keys, output paths, or media strategy values.
    They must come from the `awesome_webpage` config.
    A missing OpenRouter model value means CONFIG_NEEDED for that generator; it does not mean
    the user does not want that media modality. Do not mark audio or video as unwanted only
    because its configured model is null.
    
    Resolved awesome_webpage config for this run:
    provider: openrouter
    openrouter.models.page_generation: {{ inputs.get('config', {}).get('awesome_webpage', {}).get('openrouter', {}).get('models', {}).get('page_generation', 'moonshotai/kimi-k2.6') }}
    openrouter.models.image_generation: {{ inputs.get('config', {}).get('awesome_webpage', {}).get('openrouter', {}).get('models', {}).get('image_generation', 'google/gemini-3-pro-image-preview') }}
    openrouter.models.audio_generation: {{ inputs.get('config', {}).get('awesome_webpage', {}).get('openrouter', {}).get('models', {}).get('audio_generation', 'openai/gpt-audio-mini') }}
    openrouter.models.video_generation: {{ inputs.get('config', {}).get('awesome_webpage', {}).get('openrouter', {}).get('models', {}).get('video_generation', 'bytedance/seedance-2.0-fast') }}
    output_dir: {{ inputs.get('config', {}).get('awesome_webpage', {}).get('output_dir') or (inputs.workspace_dir ~ '/awesome-webpage-output') }}
    This resolved output_dir is configured and must not be reported as CONFIG_NEEDED.
    media_strategy.search_first: {{ inputs.get('config', {}).get('awesome_webpage', {}).get('media_strategy', {}).get('search_first', true) }}
    media_strategy.search_modalities: {{ inputs.get('config', {}).get('awesome_webpage', {}).get('media_strategy', {}).get('search_modalities', ['images']) | join(', ') }}
    media_strategy.direct_aigc_modalities: {{ inputs.get('config', {}).get('awesome_webpage', {}).get('media_strategy', {}).get('direct_aigc_modalities', ['audio', 'video']) | join(', ') }}
    media_strategy.aigc_fallback_when_search_empty: {{ inputs.get('config', {}).get('awesome_webpage', {}).get('media_strategy', {}).get('aigc_fallback_when_search_empty', true) }}
    media_strategy.allow_remote_embeds: {{ inputs.get('config', {}).get('awesome_webpage', {}).get('media_strategy', {}).get('allow_remote_embeds', false) }}
    
    User request:
    {{ inputs.user_message | xml_escape | truncate(1600) }}
    

  • id: project_slug kind: llm_chat depends_on: [requirement_framing] with: system: | You produce a single filesystem-safe slug identifying this webpage project. Do not call tools. task: | Output ONLY a single slug derived from the topic in the framed requirements below. Constraints: - lowercase ASCII letters, digits, and hyphens only - no spaces, no path separators, no file extension, no quotes - max 40 characters - topic-derived, so different topics produce different slugs

---

AwesomeWebpageMetaSkill

Build a local multimedia webpage project from a topic, audience, language, style, and media preference request.

Pipeline:

1. Requirement Framing
2. Deep Research
3. Page Outline (section structure + media intents; no filenames)
4. Media Acquisition (search / AIGC; producers emit *_READY: lines)
5. Media Assets Collect (deterministic path + existence check)
6. Webpage Source Generation
7. Deterministic Webpage Write
8. Deterministic Media Bind + Validation
9. Local Validation
10. Publish Complete Webpage Bundle
11. Delivery Guide

The current code-owned media capability candidate is OpenRouter. Its volatile credential, endpoint, and proxy are leased from ordinary Provider Settings only after the explicit media-send approval gate; they never enter this plan or a persisted run. Non-secret media model defaults, output directory, and media strategy remain configuration-owned, and individual steps must not invent them.