Ruwana

Read one owner-bound Ruwana production job.

ruwana_job_status

Ruwana Avatar has two exclusive execution paths. Script/TTS path: provide script + one exact Ruwana Avatar voice + one exact Avatar emotion + speech rate 70-130 + performance. Uploaded-audio path: provide audio + performance only; do not also send script, voice, emotion or speech-rate controls. Performance is the Avatar PERFORMANCE / CAMERA direction and stays separate from voice controls. The complete contract is validated before media transfer/provider execution.

ruwana_avatar_generate

Ruwana LAB has three distinct contracts. Creative is a clean GPT Image 2 workspace: write the actual user/creative instruction directly and do not invent an MCP hidden rewrite. Product is dedicated controlled commercial product presentation and keeps Ruwana's existing internal Product mode logic; an optional prompt may add the requested commercial setting or packaging presentation. Polish is direct cleanup/restoration and keeps Ruwana's existing internal Polish mode logic; do not send a transformation prompt. The server validates the selected LAB mode before any ChatGPT file transfer.

ruwana_lab_generate

Ruwana Motion transfers the movement from the Motion Source video to the Character image. It is not general Video and must not invent a shot plan. Choose character_orientation from the real intent: video follows the source-video movement/orientation and supports complex motion; image prioritizes the character-image composition. performance is optional and, when useful, must stay short and describe only performance interpretation such as confident, elegant, energetic, restrained or comic plus natural expression/fabric behavior. Do not put camera, lens, orbit, zoom or scene-direction instructions into performance; those are already carried by the motion source. Runtime duration is derived server-side from the real source video.

ruwana_motion_generate

Ruwana PRO photographic production. Before execution, complete the mandatory Tiger-derived direction contract: commercial objective, one role/fidelity instruction for every reference, scene/action, styling/beauty, model performance, lighting/atmosphere, composition/finish and camera intent. The server compiles that structured direction into the final PRO prompt. For a genuinely new pose, frame or shot, rebuild from the original required inputs rather than feeding a previous PRO output back merely to change pose/frame.

ruwana_pro_generate

Ruwana Video direction and execution. Before execution, complete the mandatory Tiger-derived Video contract: concrete action, continuity, lighting, one explicit camera plan and real audio intent. Single mode requires one direction block plus one camera and one continuous evolution. Slots mode requires 2-6 complete shot blocks, each with its own action, lighting, duration and camera; continuity is included only where it matters. The server compiles those fields into the existing Video prompt before media transfer. Use start_result_id for an owned canonical Ruwana image instead of download/re-upload. Do not route simple Quick Animate requests here.

ruwana_video_generate

Render a ready Ruwana result already returned by a Ruwana production, Job Status, Recent Results, or Favorites tool. Always call the relevant Ruwana data/production tool first and pass its exact result_id, module/status and media. This tool is presentation-only: displaying an image does not upload it into ChatGPT visual context or set imageIds. Use ruwana_inspect_result_image separately only when actual pixel inspection is required. It never generates, mutates, bills, or retries production.

ruwana_display_result

Return the exact canonical MP4 for one authenticated-user READY Ruwana Video, Motion, or Avatar result. Use this only when the user asks you to inspect, judge, explain, diagnose, compare, or correct what is actually in the video (for example: 'what is this?', 'this is not what I wanted', 'what do you see in it?', or 'where did it go wrong?'). Do not call it merely to show/display a result; ruwana_display_result remains the preview/player path. The file is read-only and does not generate, mutate, bill, favorite, or duplicate Ruwana media.

ruwana_get_result_file

Prepare one already-resolved READY Ruwana image as ChatGPT visual context only when actual pixel inspection is needed: describe, judge, compare, diagnose, correct, or author downstream direction that depends on what is visibly in the frame. Do not call this for simple display, status, replay, download, Favorite, or Quick Animate when canonical result identity alone is sufficient. This is read-only and does not generate, mutate, bill, favorite, or replace canonical Ruwana media.

ruwana_inspect_result_image

List Ruwana PRO virtual models available to the authenticated Ruwana user.

ruwana_pro_models