← WidemileCONTENT HISTORY

Update to Widemile

Snapshot Oct 29, 2026 · 00:05 UTC · version 1.2.0

Collection source: downloaded plugin package.

WHAT CHANGED

Payment or plan references changed

Skill instructions — Now says: the dedicated Editor operation tools when assembly is needed. - **Subtitles only when requested:** `prostudio_video_edit_prepare` and `prostudio_video_edit`...

Observed in published text. Live prices and checkout terms have not been verified by this change.

Save Widemile and check future changes in Following.

Up to 3 plugins free · no card required.

All fields and original values.

Skill instructions

Before

Oct 24, 2026 · 13:07 UTC

`prostudio_editor_apply` when assembly is needed. - **Subtitles only when requested:** `prostudio_video_edit_prepare` and `prostudio_video_edit` for automatic styled captions on a completed voiced video; `prostudio_editor_get` and `prost...

After

Oct 29, 2026 · 00:05 UTC

the dedicated Editor operation tools when assembly is needed. - **Subtitles only when requested:** `prostudio_video_edit_prepare` and `prostudio_video_edit` for automatic styled captions on a completed voiced video; `prostudio_editor_get...

Compare saved observations

Download comparison JSON

SKILL.md text changes

− removed · + added

--- before
+++ after
@@ -9,8 +9,8 @@
 
 - **Script and Brief:** `prostudio_brief_get`, `prostudio_brief_replace`, and `prostudio_workspace_get`.
 - **Optional assets and clip continuity:** `prostudio_references_list`, `prostudio_references_resolve`, `prostudio_reference_create`, and `inspect_prostudio_media` when media is supplied or a completed clip becomes the next reference.
-- **Generation and delivery:** `prostudio_models_list`, `prostudio_generation_prepare`, `prostudio_video_generate`, `prostudio_job_status`, `prostudio_generation_get`, `prostudio_editor_get`, and `prostudio_editor_apply` when assembly is needed.
-- **Subtitles only when requested:** `prostudio_video_edit_prepare` and `prostudio_video_edit` for automatic styled captions on a completed voiced video; `prostudio_editor_get` and `prostudio_editor_apply` for timed text on the assembled sequence.
+- **Generation and delivery:** `prostudio_models_list`, `prostudio_generation_prepare`, `prostudio_video_generate`, `prostudio_job_status`, `prostudio_generation_get`, `prostudio_editor_get`, and the dedicated Editor operation tools when assembly is needed.
+- **Subtitles only when requested:** `prostudio_video_edit_prepare` and `prostudio_video_edit` for automatic styled captions on a completed voiced video; `prostudio_editor_get` and the dedicated Editor operation tools for timed text on the assembled sequence.
 - This declaration does not activate tools. Use only the tools required by the selected workflow. If a required tool is absent from the current scoped catalog and `load_prostudio_tools` is available, call it with the missing outcome before claiming the capability unavailable. A prepared plan alone is not a finished explainer video.
 
 Create a finished **faceless explainer video about the user's topic**. A topic or idea alone is enough to begin. Images, footage, logos, colors, fonts, and other brand assets are optional; use supplied assets accurately when present, but do not require or invent them.
@@ -34,7 +34,7 @@
 - Generate the first clip from the selected script and any optional user assets. After it completes, inspect its actual output, resolve its durable video, and register or reuse that **exact completed video** through `reference-management`. Bind it explicitly as a video reference for the next clip. In **each successor's video-generation prompt**, require the new clip to begin as the immediate temporal continuation of the reference video's **final seconds**: match the last visible composition, subject and object positions, camera movement, lighting, and ongoing motion, then advance into the next scripted beat. Do not replay the reference's opening or restart the scene. Preserve the **user-selected style bible**, including palette, materials or illustration treatment, typography treatment, and motion rhythm. Inspect the actual seam between clips; if the model repeats the ending, trim only the duplicated overlap in the Editor before assembly. Repeat with each newly completed clip as the reference for its successor; do not substitute a still frame, merely mention the previous clip in text, regenerate the same beat, or proceed without the binding. If a later model cannot accept the clip, explain the constraint and let the user choose a compatible model before continuing.
 - If voiceover is requested, choose a video model with suitable native audio and put that scene's exact approved narration directly into **each video-generation prompt**, with the language, consistent voice direction, and its timing relative to the chosen animation. Ask for new narration only in each successive clip so the previous reference video's audio is not repeated. Verify the actual spoken output before calling a scene complete; do not imply the model spoke the approved words when it did not. If voiceover is declined, do not ask video models to speak the script.
 - Do not promise exact spelling or legible text from a generative video model. Use editable text overlays for exact on-screen copy when the available editor supports them, or explain that limitation before generation.
-- Tell the user that **subtitles are a post-production step**, separate from the generated scene graphics. Add them only if requested, after the voiceover and clip timing are settled; do not burn subtitle requests into video-generation prompts. For a completed voiced video version, load `editor-workflows`, call `prostudio_video_edit_prepare` with `edits.captions` and the desired style, then execute the exact approved `prostudio_video_edit` action after transcription is ready. For the assembled Editor timeline, use `prostudio_editor_get` and `prostudio_editor_apply` to place timed text clips against the final audio. If there is no spoken audio, create requested text cards or timed copy manually rather than claiming automatic speech captions. Check spelling, language, timing, and readability against the approved script.
+- Tell the user that **subtitles are a post-production step**, separate from the generated scene graphics. Add them only if requested, after the voiceover and clip timing are settled; do not burn subtitle requests into video-generation prompts. For a completed voiced video version, load `editor-workflows`, call `prostudio_video_edit_prepare` with `edits.captions` and the desired style, then execute the exact approved `prostudio_video_edit` action after transcription is ready. For the assembled Editor timeline, use `prostudio_editor_get` and the dedicated Editor operation tools to place timed text clips against the final audio. If there is no spoken audio, create requested text cards or timed copy manually rather than claiming automatic speech captions. Check spelling, language, timing, and readability against the approved script.
 - Before paid generation, show the saved script/storyboard, selected model or models, target final duration, planned clip durations, reference strategy, aspect ratio, audio plan, and exact known credit charge from prepared plans. Later clips can be prepared precisely only after their predecessor exists. Follow the host's approval requirement for each prepared action, then execute without changing its parameters. Inspect the assembled video and fix only visible problems that prevent the idea from reading clearly.
 
 ## Review
@@ -46,3 +46,5 @@
 - Motion, narration, and any supplied brand assets stay consistent from opening to ending.
 - Every clip after the first binds the actual preceding completed video as its visual reference and starts from its final seconds with a visually continuous transition; the final assembly matches the approved target duration as closely as the editor allows.
 - The delivered result is a playable video, not only a script, storyboard, or unexecuted plan.
+
+- Billing display is controlled by the authenticated account. When `billingDisplay.unit` is `EUR`, present the returned euro charge and balances; otherwise present credits. Preserve prepared execution arguments and `expectedCredits` exactly, without displaying those machine values as prices.
Full technical diff · 1 changed fields

changed /skill_md_contents

BEFORE
"---\nname: motion-explainer-video\ndescription: Turn one topic into a faceless explainer video with an opening hook and a user-chosen motion style, using optional images, footage, logos, or brand assets.\n---\n\n# Explainer video\n\n## Tool requirements\n\n- **Script and Brief:** `prostudio_brief_get`, `prostudio_brief_replace`, and `prostudio_workspace_get`.\n- **Optional assets and clip continuity:** `prostudio_references_list`, `prostudio_references_resolve`, `prostudio_reference_create`, and `inspect_prostudio_media` when media is supplied or a completed clip becomes the next reference.\n- **Generation and delivery:** `prostudio_models_list`, `prostudio_generation_prepare`, `prostudio_video_generate`, `prostudio_job_status`, `prostudio_generation_get`, `prostudio_editor_get`, and `prostudio_editor_apply` when assembly is needed.\n- **Subtitles only when requested:** `prostudio_video_edit_prepare` and `prostudio_video_edit` for automatic styled captions on a completed voiced video; `prostudio_editor_get` and `prostudio_editor_apply` for timed text on the assembled sequence.\n- This declaration does not activate tools. Use only the tools required by the selected workflow. If a required tool is absent from the current scoped catalog and `load_prostudio_tools` is available, call it with the missing outcome before claiming the capability unavailable. A prepared plan alone is not a finished explainer video.\n\nCreate a finished **faceless explainer video about the user's topic**. A topic or idea alone is enough to begin. Images, footage, logos, colors, fonts, and other brand assets are optional; use supplied assets accurately when present, but do not require or invent them.\n\n## Visual direction\n\n- Open the script with a strong hook in the first few seconds: a compelling question, unexpected contrast, or visual proposition tied directly to the topic. Pay it off in the explanation; do not use an unrelated teaser.\n- **Do not assume a paper-cut style.** During ideation, ask which motion style the user wants and offer a short, topic-relevant menu such as: **paper-cut/stop-motion collage** (layered cutouts and tactile textures); **flat 2D vector animation** (clean illustrated shapes and infographics); **whiteboard/hand-drawn** (diagrams drawn as the idea unfolds); **editorial mixed media** (photos, archival imagery, maps, charts, and graphic callouts); **kinetic typography** (a few bold words animated with shapes); or **stylized 3D/isometric** (simple dimensional objects and spaces). These are starting points, not a fixed template; accept the user's own style or visual reference. Explain each option briefly in terms of how it will help this topic, then let the user choose before locking the storyboard.\n- Keep the video faceless in the explainer sense: no live presenter, talking head, or lip-synced human performance. If people are needed, depict them in a way that belongs to the chosen style, such as vector figures, drawn silhouettes, paper cutouts, or 3D figures. Marble busts or statues are an optional choice within a suitable style, never a required character treatment.\n- Define a compact style bible for the chosen look: illustration or material treatment, palette, character and object design, typography treatment, camera language, and motion rhythm. Carry it through every generated clip. In each generation prompt, restate these selected visual anchors so the beats assemble into one coherent video.\n- In **every video-generation prompt**, explicitly forbid small text, fine print, tiny labels, dense paragraphs, and decorative pseudo-writing. If words must appear in the generated scene, use only a few short text elements at medium-to-large size, with clear spacing and contrast. Small model-generated lettering tends to become hallucinated and illegible. Prefer clean shapes or surfaces without generated lettering; add exact copy as editable overlays in post-production when needed. For kinetic typography, create exact words as editable overlays rather than trusting the video model to spell or animate them accurately.\n\n## Workflow\n\n- Ask for the topic if it is missing. Identify the audience, the single main takeaway, and any stated language, target duration, format, tone, or brand constraints. **Ask explicitly which motion style the user wants**, using a few relevant options from the menu above; do not silently choose paper-cut or any other style. If the user already supplied a style or visual reference, use it without asking again. **Ask explicitly whether the user wants a voiceover.** If yes, settle its language and broad voice direction before finalizing narration; if no, write a visual-first script without spoken lines. Ask only for other choices that materially change the result; propose reasonable defaults for the rest.\n- Choose the script with the user before generation. Offer two concise hook and narrative directions when useful, then draft the chosen full script and scene-by-scene storyboard in the **selected motion style**. Begin with the hook, develop the explanation, and reach a clear ending. For each scene, specify its target timing, the point it explains, the motion and visual action, narration if used, and the exact on-screen words. Keep on-screen copy sparse and medium-to-large; move longer explanations into narration or later editable overlays. Avoid packing several claims into one shot. Include a call to action only when the user wants one.\n- Once the user has chosen or revised the script, load `production-planning`, read the current Brief, and save the selected script, scene timings, exact copy, **chosen motion style and style bible** with `prostudio_brief_replace`, preserving unrelated Brief content. Keep model names, reference IDs, generation IDs, credit costs, and job status out of the Brief. Read the saved Brief for later clip prompts so the narrative and visual style remain stable.\n- If the user supplies images or footage, inspect and bind the exact selected files through `media-inspection` and `reference-management`. Preserve recognizable products, people, and brand marks. If the user supplies no media, design the explainer from generated visuals and graphics without requesting placeholder assets.\n- Load `generation-planning` and inspect the live video-model catalog. Consider **Gemini Omni Flash Lite** or **MiniMax H3 Max** according to the chosen style, reference-video support, audio language, and current capabilities; MiniMax H3 Max may be preferable when visible text or non-English speech is especially important. Verify that the selected live mode accepts the preceding video as a reference for every clip after the first. Honor an explicit user model choice, and explain any incompatibility with the required video-reference chain before preparing a plan.\n- Work backward from the user's requested final duration. Allocate timed scenes and supported clip durations, accounting for transitions and any trims in the Editor. Before generating a clip, verify that its duration and format will be accepted as a video reference by the model planned for the following clip. A long explainer is several approved clips assembled with `editor-workflows`, not one generation with an unsupported duration.\n- Generate the first clip from the selected script and any optional user assets. After it completes, inspect its actual output, resolve its durable video, and register or reuse that **exact completed video** through `reference-management`. Bind it explicitly as a video reference for the next clip. In **each successor's video-generation prompt**, require the new clip to begin as the immediate temporal continuation of the reference video's **final seconds**: match the last visible composition, subject and object positions, camera movement, lighting, and ongoing motion, then advance into the next scripted beat. Do not replay the reference's opening or restart the scene. Preserve the **user-selected style bible**, including palette, materials or illustration treatment, typography treatment, and motion rhythm. Inspect the actual seam between clips; if the model repeats the ending, trim only the duplicated overlap in the Editor before assembly. Repeat with each newly completed clip as the reference for its successor; do not substitute a still frame, merely mention the previous clip in text, regenerate the same beat, or proceed without the binding. If a later model cannot accept the clip, explain the constraint and let the user choose a compatible model before continuing.\n- If voiceover is requested, choose a video model with suitable native audio and put that scene's exact approved narration directly into **each video-generation prompt**, with the language, consistent voice direction, and its timing relative to the chosen animation. Ask for new narration only in each successive clip so the previous reference video's audio is not repeated. Verify the actual spoken output before calling a scene complete; do not imply the model spoke the approved words when it did not. If voiceover is declined, do not ask video models to speak the script.\n- Do not promise exact spelling or legible text from a generative video model. Use editable text overlays for exact on-screen copy when the available editor supports them, or explain that limitation before generation.\n- Tell the user that **subtitles are a post-production step**, separate from the generated scene graphics. Add them only if requested, after the voiceover and clip timing are settled; do not burn subtitle requests into video-generation prompts. For a completed voiced video version, load `editor-workflows`, call `prostudio_video_edit_prepare` with `edits.captions` and the desired style, then execute the exact approved `prostudio_video_edit` action after transcription is ready. For the assembled Editor timeline, use `prostudio_editor_get` and `prostudio_editor_apply` to place timed text clips against the final audio. If there is no spoken audio, create requested text cards or timed copy manually rather than claiming automatic speech captions. Check spelling, language, timing, and readability against the approved script.\n- Before paid generation, show the saved script/storyboard, selected model or models, target final duration, planned clip durations, reference strategy, aspect ratio, audio plan, and exact known credit charge from prepared plans. Later clips can be prepared precisely only after their predecessor exists. Follow the host's approval requirement for each prepared action, then execute without changing its parameters. Inspect the assembled video and fix only visible problems that prevent the idea from reading clearly.\n\n## Review\n\n- The core message is understandable without prior context, and each scene advances it.\n- The opening hook is immediate and relevant; the user's chosen motion style remains consistent throughout.\n- Generated scenes contain no small text or dense lettering; any on-screen words are few, medium-to-large, readable, and remain long enough to follow.\n- The voiceover choice is reflected in the final audio; requested subtitles are added and checked in post-production, not embedded in generated video frames.\n- Motion, narration, and any supplied brand assets stay consistent from opening to ending.\n- Every clip after the first binds the actual preceding completed video as its visual reference and starts from its final seconds with a visually continuous transition; the final assembly matches the approved target duration as closely as the editor allows.\n- The delivered result is a playable video, not only a script, storyboard, or unexecuted plan.\n"
AFTER
"---\nname: motion-explainer-video\ndescription: Turn one topic into a faceless explainer video with an opening hook and a user-chosen motion style, using optional images, footage, logos, or brand assets.\n---\n\n# Explainer video\n\n## Tool requirements\n\n- **Script and Brief:** `prostudio_brief_get`, `prostudio_brief_replace`, and `prostudio_workspace_get`.\n- **Optional assets and clip continuity:** `prostudio_references_list`, `prostudio_references_resolve`, `prostudio_reference_create`, and `inspect_prostudio_media` when media is supplied or a completed clip becomes the next reference.\n- **Generation and delivery:** `prostudio_models_list`, `prostudio_generation_prepare`, `prostudio_video_generate`, `prostudio_job_status`, `prostudio_generation_get`, `prostudio_editor_get`, and the dedicated Editor operation tools when assembly is needed.\n- **Subtitles only when requested:** `prostudio_video_edit_prepare` and `prostudio_video_edit` for automatic styled captions on a completed voiced video; `prostudio_editor_get` and the dedicated Editor operation tools for timed text on the assembled sequence.\n- This declaration does not activate tools. Use only the tools required by the selected workflow. If a required tool is absent from the current scoped catalog and `load_prostudio_tools` is available, call it with the missing outcome before claiming the capability unavailable. A prepared plan alone is not a finished explainer video.\n\nCreate a finished **faceless explainer video about the user's topic**. A topic or idea alone is enough to begin. Images, footage, logos, colors, fonts, and other brand assets are optional; use supplied assets accurately when present, but do not require or invent them.\n\n## Visual direction\n\n- Open the script with a strong hook in the first few seconds: a compelling question, unexpected contrast, or visual proposition tied directly to the topic. Pay it off in the explanation; do not use an unrelated teaser.\n- **Do not assume a paper-cut style.** During ideation, ask which motion style the user wants and offer a short, topic-relevant menu such as: **paper-cut/stop-motion collage** (layered cutouts and tactile textures); **flat 2D vector animation** (clean illustrated shapes and infographics); **whiteboard/hand-drawn** (diagrams drawn as the idea unfolds); **editorial mixed media** (photos, archival imagery, maps, charts, and graphic callouts); **kinetic typography** (a few bold words animated with shapes); or **stylized 3D/isometric** (simple dimensional objects and spaces). These are starting points, not a fixed template; accept the user's own style or visual reference. Explain each option briefly in terms of how it will help this topic, then let the user choose before locking the storyboard.\n- Keep the video faceless in the explainer sense: no live presenter, talking head, or lip-synced human performance. If people are needed, depict them in a way that belongs to the chosen style, such as vector figures, drawn silhouettes, paper cutouts, or 3D figures. Marble busts or statues are an optional choice within a suitable style, never a required character treatment.\n- Define a compact style bible for the chosen look: illustration or material treatment, palette, character and object design, typography treatment, camera language, and motion rhythm. Carry it through every generated clip. In each generation prompt, restate these selected visual anchors so the beats assemble into one coherent video.\n- In **every video-generation prompt**, explicitly forbid small text, fine print, tiny labels, dense paragraphs, and decorative pseudo-writing. If words must appear in the generated scene, use only a few short text elements at medium-to-large size, with clear spacing and contrast. Small model-generated lettering tends to become hallucinated and illegible. Prefer clean shapes or surfaces without generated lettering; add exact copy as editable overlays in post-production when needed. For kinetic typography, create exact words as editable overlays rather than trusting the video model to spell or animate them accurately.\n\n## Workflow\n\n- Ask for the topic if it is missing. Identify the audience, the single main takeaway, and any stated language, target duration, format, tone, or brand constraints. **Ask explicitly which motion style the user wants**, using a few relevant options from the menu above; do not silently choose paper-cut or any other style. If the user already supplied a style or visual reference, use it without asking again. **Ask explicitly whether the user wants a voiceover.** If yes, settle its language and broad voice direction before finalizing narration; if no, write a visual-first script without spoken lines. Ask only for other choices that materially change the result; propose reasonable defaults for the rest.\n- Choose the script with the user before generation. Offer two concise hook and narrative directions when useful, then draft the chosen full script and scene-by-scene storyboard in the **selected motion style**. Begin with the hook, develop the explanation, and reach a clear ending. For each scene, specify its target timing, the point it explains, the motion and visual action, narration if used, and the exact on-screen words. Keep on-screen copy sparse and medium-to-large; move longer explanations into narration or later editable overlays. Avoid packing several claims into one shot. Include a call to action only when the user wants one.\n- Once the user has chosen or revised the script, load `production-planning`, read the current Brief, and save the selected script, scene timings, exact copy, **chosen motion style and style bible** with `prostudio_brief_replace`, preserving unrelated Brief content. Keep model names, reference IDs, generation IDs, credit costs, and job status out of the Brief. Read the saved Brief for later clip prompts so the narrative and visual style remain stable.\n- If the user supplies images or footage, inspect and bind the exact selected files through `media-inspection` and `reference-management`. Preserve recognizable products, people, and brand marks. If the user supplies no media, design the explainer from generated visuals and graphics without requesting placeholder assets.\n- Load `generation-planning` and inspect the live video-model catalog. Consider **Gemini Omni Flash Lite** or **MiniMax H3 Max** according to the chosen style, reference-video support, audio language, and current capabilities; MiniMax H3 Max may be preferable when visible text or non-English speech is especially important. Verify that the selected live mode accepts the preceding video as a reference for every clip after the first. Honor an explicit user model choice, and explain any incompatibility with the required video-reference chain before preparing a plan.\n- Work backward from the user's requested final duration. Allocate timed scenes and supported clip durations, accounting for transitions and any trims in the Editor. Before generating a clip, verify that its duration and format will be accepted as a video reference by the model planned for the following clip. A long explainer is several approved clips assembled with `editor-workflows`, not one generation with an unsupported duration.\n- Generate the first clip from the selected script and any optional user assets. After it completes, inspect its actual output, resolve its durable video, and register or reuse that **exact completed video** through `reference-management`. Bind it explicitly as a video reference for the next clip. In **each successor's video-generation prompt**, require the new clip to begin as the immediate temporal continuation of the reference video's **final seconds**: match the last visible composition, subject and object positions, camera movement, lighting, and ongoing motion, then advance into the next scripted beat. Do not replay the reference's opening or restart the scene. Preserve the **user-selected style bible**, including palette, materials or illustration treatment, typography treatment, and motion rhythm. Inspect the actual seam between clips; if the model repeats the ending, trim only the duplicated overlap in the Editor before assembly. Repeat with each newly completed clip as the reference for its successor; do not substitute a still frame, merely mention the previous clip in text, regenerate the same beat, or proceed without the binding. If a later model cannot accept the clip, explain the constraint and let the user choose a compatible model before continuing.\n- If voiceover is requested, choose a video model with suitable native audio and put that scene's exact approved narration directly into **each video-generation prompt**, with the language, consistent voice direction, and its timing relative to the chosen animation. Ask for new narration only in each successive clip so the previous reference video's audio is not repeated. Verify the actual spoken output before calling a scene complete; do not imply the model spoke the approved words when it did not. If voiceover is declined, do not ask video models to speak the script.\n- Do not promise exact spelling or legible text from a generative video model. Use editable text overlays for exact on-screen copy when the available editor supports them, or explain that limitation before generation.\n- Tell the user that **subtitles are a post-production step**, separate from the generated scene graphics. Add them only if requested, after the voiceover and clip timing are settled; do not burn subtitle requests into video-generation prompts. For a completed voiced video version, load `editor-workflows`, call `prostudio_video_edit_prepare` with `edits.captions` and the desired style, then execute the exact approved `prostudio_video_edit` action after transcription is ready. For the assembled Editor timeline, use `prostudio_editor_get` and the dedicated Editor operation tools to place timed text clips against the final audio. If there is no spoken audio, create requested text cards or timed copy manually rather than claiming automatic speech captions. Check spelling, language, timing, and readability against the approved script.\n- Before paid generation, show the saved script/storyboard, selected model or models, target final duration, planned clip durations, reference strategy, aspect ratio, audio plan, and exact known credit charge from prepared plans. Later clips can be prepared precisely only after their predecessor exists. Follow the host's approval requirement for each prepared action, then execute without changing its parameters. Inspect the assembled video and fix only visible problems that prevent the idea from reading clearly.\n\n## Review\n\n- The core message is understandable without prior context, and each scene advances it.\n- The opening hook is immediate and relevant; the user's chosen motion style remains consistent throughout.\n- Generated scenes contain no small text or dense lettering; any on-screen words are few, medium-to-large, readable, and remain long enough to follow.\n- The voiceover choice is reflected in the final audio; requested subtitles are added and checked in post-production, not embedded in generated video frames.\n- Motion, narration, and any supplied brand assets stay consistent from opening to ending.\n- Every clip after the first binds the actual preceding completed video as its visual reference and starts from its final seconds with a visually continuous transition; the final assembly matches the approved target duration as closely as the editor allows.\n- The delivered result is a playable video, not only a script, storyboard, or unexecuted plan.\n\n- Billing display is controlled by the authenticated account. When `billingDisplay.unit` is `EUR`, present the returned euro charge and balances; otherwise present credits. Preserve prepared execution arguments and `expectedCredits` exactly, without displaying those machine values as prices.\n"
Full snapshot data
{
  "description": "Turn one topic into a faceless explainer video with an opening hook and a user-chosen motion style, using optional images, footage, logos, or brand assets.",
  "included_files": [],
  "name": "motion-explainer-video",
  "skill_md_contents": "---\nname: motion-explainer-video\ndescription: Turn one topic into a faceless explainer video with an opening hook and a user-chosen motion style, using optional images, footage, logos, or brand assets.\n---\n\n# Explainer video\n\n## Tool requirements\n\n- **Script and Brief:** `prostudio_brief_get`, `prostudio_brief_replace`, and `prostudio_workspace_get`.\n- **Optional assets and clip continuity:** `prostudio_references_list`, `prostudio_references_resolve`, `prostudio_reference_create`, and `inspect_prostudio_media` when media is supplied or a completed clip becomes the next reference.\n- **Generation and delivery:** `prostudio_models_list`, `prostudio_generation_prepare`, `prostudio_video_generate`, `prostudio_job_status`, `prostudio_generation_get`, `prostudio_editor_get`, and the dedicated Editor operation tools when assembly is needed.\n- **Subtitles only when requested:** `prostudio_video_edit_prepare` and `prostudio_video_edit` for automatic styled captions on a completed voiced video; `prostudio_editor_get` and the dedicated Editor operation tools for timed text on the assembled sequence.\n- This declaration does not activate tools. Use only the tools required by the selected workflow. If a required tool is absent from the current scoped catalog and `load_prostudio_tools` is available, call it with the missing outcome before claiming the capability unavailable. A prepared plan alone is not a finished explainer video.\n\nCreate a finished **faceless explainer video about the user's topic**. A topic or idea alone is enough to begin. Images, footage, logos, colors, fonts, and other brand assets are optional; use supplied assets accurately when present, but do not require or invent them.\n\n## Visual direction\n\n- Open the script with a strong hook in the first few seconds: a compelling question, unexpected contrast, or visual proposition tied directly to the topic. Pay it off in the explanation; do not use an unrelated teaser.\n- **Do not assume a paper-cut style.** During ideation, ask which motion style the user wants and offer a short, topic-relevant menu such as: **paper-cut/stop-motion collage** (layered cutouts and tactile textures); **flat 2D vector animation** (clean illustrated shapes and infographics); **whiteboard/hand-drawn** (diagrams drawn as the idea unfolds); **editorial mixed media** (photos, archival imagery, maps, charts, and graphic callouts); **kinetic typography** (a few bold words animated with shapes); or **stylized 3D/isometric** (simple dimensional objects and spaces). These are starting points, not a fixed template; accept the user's own style or visual reference. Explain each option briefly in terms of how it will help this topic, then let the user choose before locking the storyboard.\n- Keep the video faceless in the explainer sense: no live presenter, talking head, or lip-synced human performance. If people are needed, depict them in a way that belongs to the chosen style, such as vector figures, drawn silhouettes, paper cutouts, or 3D figures. Marble busts or statues are an optional choice within a suitable style, never a required character treatment.\n- Define a compact style bible for the chosen look: illustration or material treatment, palette, character and object design, typography treatment, camera language, and motion rhythm. Carry it through every generated clip. In each generation prompt, restate these selected visual anchors so the beats assemble into one coherent video.\n- In **every video-generation prompt**, explicitly forbid small text, fine print, tiny labels, dense paragraphs, and decorative pseudo-writing. If words must appear in the generated scene, use only a few short text elements at medium-to-large size, with clear spacing and contrast. Small model-generated lettering tends to become hallucinated and illegible. Prefer clean shapes or surfaces without generated lettering; add exact copy as editable overlays in post-production when needed. For kinetic typography, create exact words as editable overlays rather than trusting the video model to spell or animate them accurately.\n\n## Workflow\n\n- Ask for the topic if it is missing. Identify the audience, the single main takeaway, and any stated language, target duration, format, tone, or brand constraints. **Ask explicitly which motion style the user wants**, using a few relevant options from the menu above; do not silently choose paper-cut or any other style. If the user already supplied a style or visual reference, use it without asking again. **Ask explicitly whether the user wants a voiceover.** If yes, settle its language and broad voice direction before finalizing narration; if no, write a visual-first script without spoken lines. Ask only for other choices that materially change the result; propose reasonable defaults for the rest.\n- Choose the script with the user before generation. Offer two concise hook and narrative directions when useful, then draft the chosen full script and scene-by-scene storyboard in the **selected motion style**. Begin with the hook, develop the explanation, and reach a clear ending. For each scene, specify its target timing, the point it explains, the motion and visual action, narration if used, and the exact on-screen words. Keep on-screen copy sparse and medium-to-large; move longer explanations into narration or later editable overlays. Avoid packing several claims into one shot. Include a call to action only when the user wants one.\n- Once the user has chosen or revised the script, load `production-planning`, read the current Brief, and save the selected script, scene timings, exact copy, **chosen motion style and style bible** with `prostudio_brief_replace`, preserving unrelated Brief content. Keep model names, reference IDs, generation IDs, credit costs, and job status out of the Brief. Read the saved Brief for later clip prompts so the narrative and visual style remain stable.\n- If the user supplies images or footage, inspect and bind the exact selected files through `media-inspection` and `reference-management`. Preserve recognizable products, people, and brand marks. If the user supplies no media, design the explainer from generated visuals and graphics without requesting placeholder assets.\n- Load `generation-planning` and inspect the live video-model catalog. Consider **Gemini Omni Flash Lite** or **MiniMax H3 Max** according to the chosen style, reference-video support, audio language, and current capabilities; MiniMax H3 Max may be preferable when visible text or non-English speech is especially important. Verify that the selected live mode accepts the preceding video as a reference for every clip after the first. Honor an explicit user model choice, and explain any incompatibility with the required video-reference chain before preparing a plan.\n- Work backward from the user's requested final duration. Allocate timed scenes and supported clip durations, accounting for transitions and any trims in the Editor. Before generating a clip, verify that its duration and format will be accepted as a video reference by the model planned for the following clip. A long explainer is several approved clips assembled with `editor-workflows`, not one generation with an unsupported duration.\n- Generate the first clip from the selected script and any optional user assets. After it completes, inspect its actual output, resolve its durable video, and register or reuse that **exact completed video** through `reference-management`. Bind it explicitly as a video reference for the next clip. In **each successor's video-generation prompt**, require the new clip to begin as the immediate temporal continuation of the reference video's **final seconds**: match the last visible composition, subject and object positions, camera movement, lighting, and ongoing motion, then advance into the next scripted beat. Do not replay the reference's opening or restart the scene. Preserve the **user-selected style bible**, including palette, materials or illustration treatment, typography treatment, and motion rhythm. Inspect the actual seam between clips; if the model repeats the ending, trim only the duplicated overlap in the Editor before assembly. Repeat with each newly completed clip as the reference for its successor; do not substitute a still frame, merely mention the previous clip in text, regenerate the same beat, or proceed without the binding. If a later model cannot accept the clip, explain the constraint and let the user choose a compatible model before continuing.\n- If voiceover is requested, choose a video model with suitable native audio and put that scene's exact approved narration directly into **each video-generation prompt**, with the language, consistent voice direction, and its timing relative to the chosen animation. Ask for new narration only in each successive clip so the previous reference video's audio is not repeated. Verify the actual spoken output before calling a scene complete; do not imply the model spoke the approved words when it did not. If voiceover is declined, do not ask video models to speak the script.\n- Do not promise exact spelling or legible text from a generative video model. Use editable text overlays for exact on-screen copy when the available editor supports them, or explain that limitation before generation.\n- Tell the user that **subtitles are a post-production step**, separate from the generated scene graphics. Add them only if requested, after the voiceover and clip timing are settled; do not burn subtitle requests into video-generation prompts. For a completed voiced video version, load `editor-workflows`, call `prostudio_video_edit_prepare` with `edits.captions` and the desired style, then execute the exact approved `prostudio_video_edit` action after transcription is ready. For the assembled Editor timeline, use `prostudio_editor_get` and the dedicated Editor operation tools to place timed text clips against the final audio. If there is no spoken audio, create requested text cards or timed copy manually rather than claiming automatic speech captions. Check spelling, language, timing, and readability against the approved script.\n- Before paid generation, show the saved script/storyboard, selected model or models, target final duration, planned clip durations, reference strategy, aspect ratio, audio plan, and exact known credit charge from prepared plans. Later clips can be prepared precisely only after their predecessor exists. Follow the host's approval requirement for each prepared action, then execute without changing its parameters. Inspect the assembled video and fix only visible problems that prevent the idea from reading clearly.\n\n## Review\n\n- The core message is understandable without prior context, and each scene advances it.\n- The opening hook is immediate and relevant; the user's chosen motion style remains consistent throughout.\n- Generated scenes contain no small text or dense lettering; any on-screen words are few, medium-to-large, readable, and remain long enough to follow.\n- The voiceover choice is reflected in the final audio; requested subtitles are added and checked in post-production, not embedded in generated video frames.\n- Motion, narration, and any supplied brand assets stay consistent from opening to ending.\n- Every clip after the first binds the actual preceding completed video as its visual reference and starts from its final seconds with a visually continuous transition; the final assembly matches the approved target duration as closely as the editor allows.\n- The delivered result is a playable video, not only a script, storyboard, or unexecuted plan.\n\n- Billing display is controlled by the authenticated account. When `billingDisplay.unit` is `EUR`, present the returned euro charge and balances; otherwise present credits. Preserve prepared execution arguments and `expectedCredits` exactly, without displaying those machine values as prices.\n"
}

SHA-256 of public snapshot: a404b256db9ab381464683925fb6cf10637340e90d4cbcd2a3068b6ca9cd671a