Use a unified production process, selecting production, optimization, and manual acceptance rules by purpose, with the source using video frame extraction by default. Video serves as the motion sketch; when the user explicitly selects the image route, form a sequence using the actually generated first and last frames, stage keyframes, and two-image reference frame interpolation. The final animation is subject to game requirements and the visual baseline. Resource delivery for the current Work Item returns to the existing control plane, and no separate animation state machine is created.
Within the scope of the current Work Item, internally propose, derive, and maintain the motion timeline design record; when requirements are clear, proceed directly with prompts and real assets according to existing authorization; the user reviews the real video, each keyframe, and the complete sequence, fixes and re-verifies according to feedback, then returns to the control plane to submit evidence and state transitions. Only when gameplay events, duration, or behavior permissions are truly missing and cannot be derived should the affected actions be asked about.
Classification Entry
For each animation, first determine the purpose type, and read only the corresponding classification reference; for multi-type batches, select rules region by region, and do not load all four documents by default.
| Purpose Type | Basis for Selection | Read Reference |
|---|---|---|
| General Animation | Objects, environments, decorations, mechanical motion, and state changes | General Animation |
| Character Action Animation | Body behaviors such as idle, movement, attack, hit reaction, and jumping | Character Action Animation |
| Effect Animation | Visual feedback such as hits, explosions, smoke, light effects, and status effects | Effect Animation |
| UI Frame Animation | Interface motion effects that require frame-by-frame bitmap assets, such as button feedback, prompts, and icons | UI Frame Animation |
The purpose type does not determine the source route, behavior unit, number of generation batches, or number of runtime animation keys. Looping, one-shot, combination, and final-state holding are playback parameters shared by all types and can be used in combination according to the task. Character actions can be further selected by looping motion, active behavior, passive reaction, and continuous combination, without adding a new skill for each action. Simple programmatic UI motion effects are handled according to the existing UI/program route.
Source Route
When the user does not specify a source, use video frame extraction by default and note it in the internal motion timeline record; image recursive frame interpolation is enabled only when the user explicitly selects it or confirms a switch. Continue using the source explicitly selected for the current task; source switching is handled according to existing change boundaries and updates the internal record, and confirmed results are not discarded due to default value changes. If the video source or generation capability is temporarily unavailable, report the gap and do not automatically switch to the image route. Both routes share classification, the game timeline, and manual acceptance, and only read the details of the current source.
| Source | Applicability Basis | Read on Demand |
|---|---|---|
| Video Frame Extraction (Default) | Use real video as a continuous motion sketch; select valid intervals and retime by stage before extracting frames | Video Rhythm Organization、Extract Frames by Region |
| Image Recursive Frame Interpolation (Explicitly Selected) | The user chooses to control key poses, shapes, or transparent edges one by one; on a fixed canvas, create the first and last frames and necessary keyframes in sequence, then use two adjacent images to generate the specified frame positions. | Recursive image frame interpolation |
In the image route, "frame interpolation" is image generation with dual-image reference and specified action time points, not optical flow interpolation that guarantees geometric correctness. For cross-view, contact, occlusion, or loops with identical first and last frames, first fill in the necessary key poses, then recurse; do not infer complete complex motion from only the first and last frames.
Shared process
- Read the visual baseline, usage type reference, andcomplete motion and timeline, and establish the internal design artifact for the current animation. Clarify the source route, behavior boundaries, initial/final states, game duration, phase intervals, key poses, events, playback method, anchor points/allowed displacement, complete motion range, target display size, FPS, and background handling. For character actions, also clarify the connection with preceding and following actions, whether it loops, and the entry/exit conditions; seeConnection and loop design. Read as neededRegion planning; when batching, each region still has independent type and time mapping.
- According toaction analysis rulesandgameplay and intent design methods, internally derive behavior boundaries, phases and motion basis, connection/looping, and pacing trade-offs from the gameplay purpose, presentation intent, and existing time constraints, and record the suggested timeline, event time points, key pose intent, size reasoning, and phase-by-phase checkpoints. For video, also derive the 768p/2K region layout, number of strips, total group consumption, and recommended plan; when requirements are clear, proceed directly with reasonable parameters. Ask questions only about gameplay events, durations, or behavior permissions that are truly missing and cannot be derived, and pause only the affected actions without waiting for confirmation of the entire text design. Do not create graybox or playable geometric mockups.
- Write the formal prompt according to "shared visual and resource constraints + type rules + current animation parameters," and directly produce real candidates or previews within the existing authorization scope. For video, choose 768p or 2K according toresolution and batching comparison: first cover all actions with the minimum number of strips, then choose the one with the lowest total group consumption under that strip count; when conditions are equal, prefer 768p; for images, handle according toimage resolution conventions. On first delivery, directly show the real video candidate/preview or the real image first frame and keyframes, with brief checkpoints; do not first deliver a text analysis and wait for confirmation. When the user confirms the actual asset, they may simultaneously accept the timing expressed by that asset, without separate confirmation of a text restatement. When there is no existing generation authorization, comply with external operation boundaries and do not treat route selection as authorization.
- For the video route, receive the real video and verify the canvas, regions, and audio track, select valid intervals as needed, retime phases, and save the derived source; deliver the actual final frame-extraction input for manual user acceptance. If the original video is already suitable, directly accept the original video; extract frames only after the actual video content, regions, and time mapping have been manually accepted; SHA does not participate in acceptance. If it fails, fix and resubmit first; correction notes or text plans do not replace acceptance of a new video. For the image route, verify the actual output size, fix a unified canvas, and confirm the first and last frames; create the necessary keyframes for the same continuous behavior one by one in chronological order, compare them with the real baseline and adjacent confirmed frames, and advance to the next keyframe only after the current frame is manually confirmed. After the keyframes are complete and confirmed, recursively fill in the target frame positions according to the internally derived time points. Actual dependencies andconsistency and connection gatesCentralized maintenance; if the parent image has defects, fix them first and do not continue propagating.
- Extract frames from videos by fixed regions, and package images according to the actual source manifest; both share background processing, proportional scaling, fixed cells/anchors, independent atlases, reports, and HTML previews. Perform local frame fixes based on feedback for the current type, preserve existing retouching results and source identity, and update the atlas, report, preview, and SHA. Defect fixes do not replace the entire source group by default.
- Use the actual complete frame sequence to manually inspect, according to the stage-by-stage checkpoints in the internal design artifact, the expressive intent, rhythm, continuity, events, single/loop/terminal states, and external action transitions, and record specific deviations and user conclusions. For local corrections with the same goal, directly rework and show before-and-after comparisons; only ask when the gameplay semantics or source route substantially changes, and do not re-request existing generation authorization. Save the classification rules, gameplay and intent derivation, actual candidate review, time mapping, and final manual acceptance evidence, according toworkflow integration contractSubmit the assets, then return to the control plane to complete Phaser integration and runtime verification.
Action design and actual asset confirmation
Action analysis and timeline are written into the internal design artifact; when requirements are clear, no separate text proposal confirmation gate is set. For the video route, the user reviews actual video candidates and the complete sequence; the user's confirmation of the actual video can simultaneously accept the timing expressed in it. For the image route, confirm the actual first and last frames, keyframes, and recursive parent images one by one, then accept the complete actual sequence. The actual sequence preview provides play/pause, replay, and frame-by-frame viewing, and check markers are placed only in the preview interface. Formal art uses the actual baseline and confirmed assets; design diagrams and historical placeholder illustrations must not be used as visual references for image generation. For local corrections with the same goal, directly rework and show before-and-after comparisons; only re-ask when the gameplay semantics or source route substantially changes. For detailed requirements, seeAction design and actual asset confirmation、Actual asset review and acceptanceandthe boundary between design and art。
Shared quality and key constraints
- Accept based on readability, rhythm, primary-secondary relationships, continuity, and event synchronization; for specific judgments and optimization methods, seeShared quality and optimization. Weight, degree of exaggeration, pauses, and follow-through motion are configured according to type, object nature, and visual baseline, and combat rhythm is not uniformly applied.
- The duration and rhythm of each animation are first determined independently based on game characteristics, gameplay response, object nature, and expressive purpose. Multiple groups of animations in the same video can have different total durations, phase proportions, pauses, loop cycles, and FPS, and are not forcibly unified for batching; for specific planning, seegame time precedes source. Game time is independent of the source video duration; the same complete behavior shares its own timeline, and events and input transitions can be marked internally; production grouping does not add attack counts, automatic combos, or damage, nor does it change existing input mechanisms.
- Duration, phase proportions, hit pauses, combo counts, and audio are all determined by the current task; specific action and parameter cases cannot become global default requirements. Videos are arranged by actual regions at 768p/2K, with the number of strips and whole-group consumption chosen accordingly, and batches are not padded by fixed tiers, compressing the motion range, or losing necessary details. Image resolution conventions remain independent. Image sequences have no video audio track, and sound is bound to events independently.
- By default, keep the required content and the complete motion range within frame, preserving safe margins. Cropping exceptions must have explicit authorization from the current task, with the object, range, and reason recorded item by item; do not crop edges on your own for batching or to emphasize force.
- A stable anchor does not mean motion is prohibited. Allowed displacement, pose exaggeration, or deformation must have its range and intent specified in the design artifact and be accepted according to plan; unplanned drift, flicker, scale changes, and cross-region intrusion are prohibited.
- A fixed canvas, identical anchor records, and passing source verification do not prove the actual object scale, support position, or motion continuity. For the image route, first confirm the type-related state of the current frame and its adjacent continuity, then release the production that depends on it; error correction cannot be replaced by adding interpolated frames, per-frame scaling and centering, or static duplication.
- Background processing is selected according to the material edges and type. Color-based background removal is only used for confirmed solid-color hard edges; semi-transparent, glowing, or unknown edges use true Alpha, masks, or frame image correction, and must not be passed off as confirmed hard edges. Whether effects, sound, and camera feedback are baked in is decided according to the task contract.
- Classification and optimization parameters are stored in the design artifact; contract fields are not added because of classification, and scripts or generation API parameters are not fabricated. Video follows the original contract; images use an independent source manifest, report, and accepted contract, recording the real generated image, parent image, time slot, generation/retouching records, and SHA, and must not forge video paths or source timestamps.
- Timeline derivation is recorded in internal artifacts; human confirmation focuses on the real video, image keyframes, and the complete sequence, while the file gate only verifies structure and identity. Automatic screenshots or launching real-device acceptance must not be performed, and external writes, payments, and publishing follow the independent authorization boundary.
Execution agent
The timeline design, prompts, time organization, frame extraction/image sequence integration, and correction of a frame animation Work Item are still uniformly delegated to gpt-6.1-sol、max 's execution sub-agent. The main agent is responsible for requirements, task packages, human feedback, integration scope, review, and final verification, and does not take over animation production.
Use collaboration.spawn_agent, set agent_type="default"、model="gpt-6.1-sol"、reasoning_effort="max"、fork_turns="none". A complete task package includes the Work Item, purpose type and source route, corresponding references, gameplay purpose and presentation intent, existing gameplay time constraints, behavior/input boundaries, stage checkpoints, material and baseline paths, allowed modification scope, current candidate version/internal timeline, allowed advancement steps, acceptance criteria, test level and commands, prohibited operations, and parallel reminders. The first delivery directly provides the real candidate/preview and a brief checkpoint, and the rationale for motion derivation is stored in internal artifacts; only gameplay conditions that cannot be derived block the affected motion. The video route hands over the current real video and human review results, and the image route hands over the currently confirmed frames and existing generation authorization to continue production.
For continuous corrections to the same work item, reuse the agent first; when the scope changes substantially or the work item is completed, end that lifecycle. Bitmap generation/editing is still handed to the image generation agent, and the animation execution agent provides the timeline, frame slots, and material constraints and integrates the results. If a specified tool or model is unavailable, report the limitation and do not silently switch models. Delegation does not expand authorization for files, automated visual verification, or external operations.
Local tools and reference routing
Video frame extraction requires ffmpeg、ffprobe and the project sharp; image planning uses only Node.js, and image packaging uses sharp. First check the actual tool help for the current route:
node <skill-dir>/scripts/video-to-atlas.mjs --help
node <skill-dir>/scripts/image-inbetween-plan.mjs --help
node <skill-dir>/scripts/images-to-atlas.mjs --help
The video tool processes one region at a time (or the entire canvas if omitted), covering the complete input time range;--fps controls both sampling and playback, and there are no time-editing or independent per-frame duration parameters. Image planning first determines the game duration, FPS, keyframe positions, and recursive dependencies; image packaging takes the FPS/loop from the source manifest, processes only actual images, and does not generate materials. For time issues, first correct the design and sources according to the shared timeline, not just the report.
| Current stage | Read on demand |
|---|---|
| Purpose determination and type creation | The corresponding classification reference above; read its rules again when independent effects or UI bitmaps are involved |
| Derive duration, pacing, and checkpoints from gameplay and presentation intent | intent-driven-design.md, read and write into existing design artifacts during motion analysis, reused for production and acceptance |
| Feedback semantics, type applicability, or event coverage rules are needed | Phaser 2D Juiciness Design Guide; this Skill still produces frame animations according to existing timeline design, real asset review, and manual acceptance of the complete sequence |
| Game time, grouping, stages, and corrections | action-timing.md |
| Effect image size inference, batching, and spatial cropping | region-planning.md |
| Image start/end, keyframes, two-image reference recursion, and image source contract | image-inbetween.md |
| Motion design and asset confirmation, prompt composition, and source/retouching evidence | workflow-integration.md |
The global stage is still controlled bythe sole control planemanagement: V2 internally derives and records the motion timeline, V3 directly produces and reviews real videos or keyframes and accepts the complete sequence, and V4 integrates into Phaser based on the report's preload, anims, and origin and verifies the runtime state.