--- name: video-director version: 1.0.0 category: 内容创作 trigger_words: tags: - image - video - ai platform: coze source: DeepseekModel source_url: https://deepseekmodel.com/skill?id=jamailar-beav-desktop-builtin-skills-video-director-skill-md --- name video-director description Canonical entrypoint for every AI chat request that asks to make, generate, plan, or edit a video, including promotional films, ads, short videos, product videos, reference-image videos, image-to-video, first/last-frame transitions, storyboard/keyframes, character talking-head videos, 宣传片, 广告片, 短视频, 视频, 口播视频, 分镜图, 图片转视频, and 商品宣传片. Do not invoke cosyvoice-ssml or tts-director as the first response to a video request. This skill must draft a script and shot table, generate storyboard contact-sheet preview images through image.generate, ask the user to confirm the script plus storyboard together, and only then call video generation. allowedRuntimeModes ["chatroom","redclaw"] allowedTools ["workflow"] activationScope session activationHint Invoke this before any media generation when the user asks for 宣传片, 广告片, 短片, 短视频, 视频, 口播视频, 数字人视频, 分镜, 分镜图, 图片转视频, reference-image video, product promotional video, character talking-head video, or an attached/reference image to become a video. In AI chat, all video-generation requests must enter this skill first. Do not call cosyvoice-ssml, tts-director, image.generate, voice.speech, or video.generate first; load this skill and follow its staged workflow. Video Director Use this skill before calling Operate(resource="video", operation="generate", input={ ... }) for video work. For AI chat, this skill is the single entrypoint for video generation. If the user says "做一个口播视频", "生成一个视频", "产品宣传片", "广告片", "数字人视频", or any similar video request, invoke video-director first. Do not start from cosyvoice-ssml ; that skill is an internal digital-human TTS substep and must not drive the video workflow. If this skill was loaded because the user attached an image and asked for a promotional film / 宣传片 / 视频, treat the attached image as a product or visual reference for the video workflow. Do not generate a replacement product image as an "analysis" step. Default Workflow Treat the workflow as a gated state machine. Do not skip or reorder states: plan_draft : Clarify the intended video mode from the user's goal and assets, then draft a concise but detailed video script and shot table. storyboard_preview : Immediately after the shot table exists, generate storyboard contact-sheet preview image(s) with Operate(resource="image", operation="generate", input={ ... }) . The storyboard image is a preview of the planned shots, not a standalone product poster, replacement product render, cover image, or final video keyframe. user_review : Show the script, storyboard preview image(s), and explicit video specs together. Ask the user to confirm or revise the whole plan. production_prep : After confirmation, decide whether this should be a <=15s single upstream clip, a long video_sequence request, or a manual editor/project assembly. Only use a video project pack if the user explicitly asks for a project/package/editor workflow, or the task is already bound to an existing pack. tts_if_needed : If the confirmed task is a character talking-head / 口播 video, complete the TTS-first workflow below. video_generate : Only after confirmation and any required TTS audio is complete, call Operate(resource="video", operation="generate", input={ ... }) . If the user has not yet confirmed the script, do not generate the video. If the user has not yet seen and approved the storyboard preview image(s), do not generate the video. If the shot table exists but no storyboard contact-sheet image has been generated yet, the next tool call must be image.generate for the storyboard preview, not video.generate , voice.speech , tts-director , or cosyvoice-ssml . The storyboard contact-sheet preview is mandatory after the first shot script is written. It is not the same as final video keyframes; it is a quick visual proof of the planned shot sequence so the user can approve direction, composition, product placement, and continuity before video generation. For product promotional videos from an attached/reference image: The first image.generate call, if any, must be for a storyboard contact sheet after the script and shot table exist. The storyboard prompt must explicitly say it is a multi-panel storyboard preview for the video, and it must preserve the attached/reference product's visible shape, material, color, and distinctive details. Do not call image.generate just to "analyze", "extract", "enhance", or recreate the product reference. Do not call video.generate until the user confirms the storyboard direction. Asset-Library Character Talking-Head Rule When the user wants a talking-head / 口播 / presenter / livestream welcome video using a character from the asset library, the character image is only the visual identity reference. It is not the speech track. Keep these three inputs separate: voiceId : the reusable cloned voice identity stored on the character asset. This is the primary voice field for future TTS calls. voice sample / reference audio: fallback source material for cloning when no ready voiceId exists. Do not keep analyzing or presenting it once a ready voiceId has been found. generated TTS audio asset: the completed spoken script. This is the only audio that should drive a character talking-head video. Use this exact order: Script and storyboard first Confirm the final spoken script. Confirm the storyboard / shot table. The Sound column must contain the actual spoken lines, not vague labels like “角色讲解”. Resolve the character voice Read the selected character asset before generating media. If the asset has a ready voiceId / voice_id , treat that as the resolved voice and stop inspecting voice samples. Report the resolved voiceId , not the reference-audio filename. If the asset has a voice sample but no ready voiceId , call Operate(resource="voice", operation="clone", input={ ... }) first, wait for completion, and use the returned voiceId . If there is no ready voiceId and no usable voice sample, stop and ask the user to provide or choose a voice. Do not fall back to a generic voice silently. Generate the complete voice track with TTS Call Operate(resource="voice", operation="speech", input={ ... }) with the full approved spoken script and the resolved voiceId . Prefer one complete audio asset for the whole talking-head segment unless the user explicitly wants separate clips. For expressive narration, long scripts, poetry, ads, or any multi-emotion performance inside this confirmed digital-human / VideoRetalk flow, activate tts-director first with Operate(resource="skills", operation="invoke", input={ "name": "tts-director" }) , then use its guidance to split the approved script into semantic beats and assign model-appropriate delivery controls. If the selected TTS model is CosyVoice, tts-director may activate cosyvoice-ssml only here, after the video-director workflow has confirmed this is a digital-human / VideoRetalk / asset-library talking-head video and the script plus role voice are resolved. Add restrained delivery controls when they improve the approved performance: use speed for pace, pitch for tone, emotion for mood, expressive punctuation such as ～ , ？ , ！ , …… , and MiniMax text markers such as <#0.6#> , (laughs) , (sighs) , or (breath) inside input only where the spoken rhythm needs them. Do not write control instructions into the spoken text. For example, pass "emotion":"happy" and speed:1.08 instead of making the character say “用开心快速的语气”. If the approved voice track needs multiple emotional beats, submit one voice.speech call with ordered segments ; the media runtime will generate each segment and merge the final audio. Do not make repeated voice.speech calls and then manually merge. Wait until the TTS result contains a usable audio asset path, URL, or asset id. Do not start video generation while the TTS job is only queued or running. Generate video with visual reference plus generated audio Use the selected character image as the visual reference. Use the newly generated complete TTS audio asset as drivingAudio . Identify it in the prompt preface as generated speech audio, for example Audio 1: generated Jamba TTS speech for lip-sync / speaking rhythm . Then call Operate(resource="video", operation="generate", input={ ... }) , normally in reference-guided mode. This is mandatory because the video model needs a finished audio waveform aligned with the approved script. A character voiceId is not audio. A voice sample is not the approved spoken line. The spoken text inside the video prompt is not a substitute for TTS. Forbidden shortcuts: Do not call video generation for character口播 with only the character image and a script. Do not pass voiceId , voice sample audio, or “音色素材” as drivingAudio . Do not rely on generateAudio to create the character voice for talking-head work. Do not describe the intended speech only in the video prompt and expect the video model to synthesize it. Do not split the TTS per storyboard row unless the user asks for separately editable audio clips or the video will be assembled from multiple independent speaking clips. Recommended TTS payload shape: { "voiceId" : "voice_xxx" , "segments" : [ { "input" : "完整、已确认的口播开场。<#0.5#>" , "speed" : 0.98 , "pitch" : 0 , "emotion" : "calm" } , { "input" : "自然地进入更有感染力的重点(laughs)。" , "speed" : 1.05 , "pitch" : 0 , "emotion" : "happy" } ] , "title" : "Jamba welcome voiceover" , "waitForCompletion" : true } Recommended video payload shape after TTS completes: { "generationMode" : "reference-guided" , "referenceImages" : [ "/absolute/path/to/jamba.png" ] , "drivingAudio" : "/absolute/path/to/generated-jamba-tts.mp3" , "drivingAudioLabel" : "Generated Jamba TTS speech from the approved script" , "storyboardShots" : [ { "time" : "0-3s" , "picture" : "Jamba faces camera in the livestream room and smiles naturally." , "sound" : "欢迎来到直播间，今天给大家准备了一个超实用的好物。" , "shot" : "Medium close-up, stable presenter framing." } ] } Video Project Pack Rule A video project pack is not the default path. Create one only when: the user explicitly asks for a 视频项目 / 视频工程 / 项目包 the user explicitly asks to continue later inside the video editor / project workbench the task is already bound to an existing video project pack Do not create one only because the request has multiple shots, long context, continuity risk, storyboard needs, or possible revisions. When it is explicitly needed, create it with: Operate(resource="video", operation="create", input={ "explicitProjectWorkflow": true, "title": "...", "duration": "...", "aspectRatio": "...", "mode": "..." }) The project folder lives in: manuscripts/video// It should be used to keep these files together: manifest.json script.md assets.json editor.project.json imported reference images / keyframes / generated clips / final output After the pack is created: write the user brief and approved script back into the video project folder keep later keyframes, clips, and outputs in the same pack whenever possible Otherwise keep the planning in chat, call video generate directly, and let the output live in the generated media library. Hard Rules Video generation is locked to the official video route configured by the app. Do not choose arbitrary video endpoints or third-party video models. Use only these official model mappings: text-to-video -> wan2.7-t2v-video reference-guided -> wan2.7-r2v-video first-last-frame -> wan2.7-i2v-video Treat first/last-frame transitions as a subtype of image-to-video work. A single upstream video generation segment must not exceed 15 seconds. For a final video longer than 15 seconds, submit one video.generate request with either the final durationSeconds or explicit videoSegments ; the media queue will generate <=15s segments and concatenate one final video asset. Do not call video.generate repeatedly just to bypass the upstream 15 second segment limit. Do not skip the script review step just because the request sounds obvious. Unless the user explicitly asks for a longer continuous shot, a single shot should usually be 1-3 seconds. Without explicit user approval, any single shot must not exceed 5 seconds. Mode Selection Use text-to-video when the user only provides text and wants a fresh video shot. Use reference-guided when the user provides one or more reference images and wants the video to absorb subject elements, style cues, props, scene motifs, or composition hints from those images. Use first-last-frame only when two images have explicit start/end semantics, such as “from A to B”, “首帧/尾帧”, “开头/结尾”, or “起始状态/结束状态”. If the user gives two images but they are only style references, do not use first-last-frame ; stay with reference-guided semantics instead. Production Strategy 单视频模式 : Use one generated video clip. Default when the request is simple, the action is short, and the full idea fits inside one coherent clip. A single generated clip must not exceed 15 seconds. 长视频队列模式 : Use this for one final AI-generated video longer than 15 seconds. Submit one video.generate request with durationSeconds > 15 ; the runtime will create a video_sequence queue job, split the generation into <=15s segments, wait for each segment, concatenate the segments, and return one final video asset. Use explicit videoSegments when the storyboard needs scene-by-scene, beat-by-beat, reference-image, first/last-frame, or driving-audio control. Each videoSegments[] item must be <=15 seconds and should correspond to a coherent group of approved storyboard rows. Keep waitForCompletion: true unless the user explicitly asks to run the long generation in the background. 手动多视频工程模式 : Use this only when the user explicitly asks for separate clips, a project/editor workflow, imported clip assembly, manual post-production, or separately editable outputs. In normal chat/agent video generation, do not manually generate one clip at a time and concatenate with ffmpeg ; use 长视频队列模式 instead. When manual assembly is explicitly needed, generate each clip deliberately, then concatenate them in storyboard order after all clips succeed. If the request has multiple shots, clear continuity requirements, or a risk of visual drift, ask one more question after showing the contact-sheet preview: whether separate storyboard keyframes should also be generated before video production. If separate storyboard keyframes are generated, later video generation should preferentially use image-based modes, and for transition-heavy segments should prefer first-last-frame . Storyboard-First Rule Every video task needs a storyboard contact-sheet preview after the script table and before user confirmation. Use generated individual keyframes in addition to the contact sheet when the request is complex enough that video quality depends on stable keyframes. Use storyboard-first when one or more of these is true: There are many shots or visual beats. Character identity must remain highly stable. Environment continuity matters. The user wants a sequence that later becomes one assembled video. The user explicitly asks for storyboard frames / keyframes / 分镜图. When any of the above is true, do not silently continue to video generation after the contact sheet. Ask the user whether they also want separate image-generated storyboard keyframes before video generation. When storyboard-first is used, follow this exact process: First design a core environment reference image . Generate that image first. Then generate later keyframes one by one. Each later keyframe must use the core environment reference image as a reference image. If a character asset already exists in the asset library, the asset reference and the core environment reference should both be preserved across later keyframes. Only after the keyframes are stable should video generation proceed. Core Environment Reference Image The first storyboard image should be a single overall environment master frame . It must contain: the full spatial layout, the key environment elements, the main subject placement, the major props, the lighting logic, the camera worldview for the sequence. This image acts as the environmental anchor for all later keyframes. Do not start by generating an isolated close-up if the later sequence depends on environment continuity. Storyboard Contact-Sheet Preview After writing the script table, generate the required storyboard effect image(s) before asking for final video approval. Use the shot count to choose the grid: 1-4 shots -> four-panel grid. 5-6 shots -> six-panel grid. 7-9 shots -> nine-panel grid. More than 9 shots -> split into multiple contact sheets; each generated image contains at most 9 storyboard panels. Rules: One generated image may contain at most 9 storyboard panels. A storyboard contact sheet is a single output image. Set count: 1 and put every panel description inside prompt . Do not use imagePlanItems for storyboard contact sheets; imagePlanItems means separate image outputs unless the tool contract explicitly says otherwise. Do not add extra invented shots to fill empty panels. Keep the panel order left-to-right, top-to-bottom, matching the script table. The preview should show composition, subject placement, camera scale, product/prop position, action beat, lighting, and style continuity. Do not render planning labels, shot numbers, table headers, or internal notes inside the image unless the user explicitly asks for visible text. The contact sheet is for visual approval only; it does not replace the approved Markdown script or storyboardShots payload used later for video generation. Reference handling is mandatory: Before any image.generate or video.generate call that depends on prior context, inspect the Current session resources block. If the needed file is not obvious there, call Operate(resource="session", operation="list", input={ "kind": "image", "limit": 20 }) or Operate(action="session.resources.list", payload={...}) and use the returned reference / path value exactly. If the user attached or selected product images, character images, brand images, scene references, or previous generated images, pass them to image.generate as referenceImages . If the reference comes from the asset library, read the asset first and pass its resolved image path(s) through referenceImages or subjectIds . If several references exist, include prompt preface lines that define each role, such as Image 1: product shape and material reference , Image 2: character identity reference , Image 3: scene mood reference . Do not describe reference images only in prose while leaving them out of the tool input. Do not invent local paths, filenames, or user folders. Use only user-provided paths, asset-library paths, or session.resources.* results. If there are more reference images than the tool supports, prioritize product/character identity first, then scene, then style. Recommended contact-sheet image.generate payload shape: { "count" : 1 , "generationMode" : "reference-guided" , "aspectRatio" : "16:9" , "prompt" : "Image 1 is the product reference: preserve shape, material, logo position, and main color. Create one cinematic storyboard contact sheet with six panels arranged left-to-right, top-to-bottom. Each panel corresponds to one storyboard row: Panel 1 visual: ... Panel 2 visual: ... Panel 3 visual: ... Panel 4 visual: ... Panel 5 visual: ... Panel 6 visual: ... No visible labels, no captions, no table text." , "referenceImages" : [ "/absolute/path/to/product.png" ] } Use text-to-image only when there are no usable visual references. Use reference-guided whenever reference images, subject assets, product photos, or prior generated images exist. Prompt Consistency Rules For Keyframe Images When using image generation to build storyboard frames, consistency matters more than flourish. You must: Define one stable description block for the subject. Define one stable description block for the environment. Reuse those same description phrases across all keyframe prompts. Only change the parts that truly differ from shot to shot. The subject anchor should usually keep these elements stable: name / identity, gender or presentation if relevant, age range if relevant, hairstyle, clothing, key facial traits, key props, visual style. The environment anchor should usually keep these elements stable: place / room type, layout, background elements, lighting mood, color palette, important objects, time-of-day logic if relevant. Do not rewrite the whole scene in a different wording for each frame. Do not keep inventing new environment details frame by frame. Do not vary the character description unless that change is intentional. Keyframe Generation Order If storyboard frames are generated: Write one explicit subject anchor block. Write one explicit environment anchor block. Generate the core environment master frame first. Generate each later keyframe individually. Each later keyframe prompt should: restate the same subject anchor, restate the same environment anchor, identify the core environment image as a reference, describe only the shot-specific difference. This is mandatory when the storyboard is later used for video generation. If those storyboard frames have already been saved into a video project pack, later video generation should use those keyframes as the main visual references. Do not keep reusing raw subject-library portraits or product stills as the primary visual input unless you truly need extra补充 angles or missing objects. Script Format The pre-generation script draft must be shown as a Markdown table. Use these columns: Time Picture Sound Shot Requirements: Before the table, explicitly state: 视频时长 视频比例 Time : use compact ranges such as 0-2s , 2-4s , 4-6s . Picture : describe subject action, motion, camera movement, scene changes, and what must stay stable. Sound : describe spoken line, ambient sound, music feel, silence, or rhythm cue.