get_project_analysis
ChatGPTUse this when the user wants the full existing analysis of one of their Viral Recreate AI projects: summary, chronological scene/shot breakdown with timestamps, image-generation fields and prompt, video-generation fields and prompt, camera/subject/object movement and sound information, image-to-video plan when the media is a still, complete image and video generation prompts, the audio-analysis flag and the output language. Read-only — it retrieves an analysis that already exists.
list_projects
ChatGPTUse this when the user wants to find, browse or list their previous Viral Recreate AI analyses — for example 'show my latest analyses', 'find the project I ran yesterday', or to look up a project id before fetching its prompts. Returns the signed-in user's projects newest first with id, name, media type, status, duration and part count. This is NOT for analyzing a new file: to analyze media, use viralrecreate_analyze_media.
rename_project
ChatGPTUse this when the user wants to give one of their Viral Recreate AI projects a clearer private name. It only changes the project's label — the uploaded media, the analysis, the prompts and all usage records stay exactly as they are, and nothing is published or deleted.
viralrecreate_analyze_media
ChatGPTUse this when the user attaches or selects an image or video in ChatGPT and wants it analyzed, turned into AI prompts, reverse-prompted, broken into scenes/shots with timestamps, described cinematographically, checked for camera/subject/object movement, transcribed or described for real sound and dialogue, converted into image prompts, video prompts, movement + sound output, complete generation prompts, or planned as an image-to-video animation from a still. Pass the attached file in the file parameter (ChatGPT supplies download_url and file_id) and Viral Recreate AI will securely ingest it into the signed-in user's account, run its ultra-detailed chronological full-timeline analysis, save it as a project, and return the real result: summary, scene-by-scene timestamps, image prompts, video prompts, movement + sound, complete image prompt, complete video prompt, language/audio flags and project id. Supported formats: JPG, PNG, WEBP, MP4, MOV, WEBM. Do not generate your own prompt instead of calling this tool. Omit file and pass project_id or project_name only to retrieve a previously saved analysis. Requires a connected Viral Recreate AI account: if the tool reports it is not connected, connect the user (log in or create a free account at viralrecreate.com) and call it again. Analyses consume the connected account's plan allowance; if it reports the allowance is used up, direct the user to ViralRecreate.com/Subscription rather than retrying.
viralrecreate_analyze_upload
ChatGPTUse this when a user wants Viral Recreate AI to analyze a newly uploaded image or video from ChatGPT and generate image prompts, video prompts, movement and sound, shot breakdowns, timestamps and complete generation prompts. Pass the temporary signed download URL, file name and MIME type of the selected file. Supported: MP4, MOV, WEBM video and JPG, PNG, WEBP images. Prefer passing the ChatGPT file object (download_url + file_id); a raw file_url is only accepted when it is an OpenAI-issued temporary file URL. The user must have a connected Viral Recreate AI account: if the tool reports it is not connected, connect the user (they can log in or create a free account at viralrecreate.com) and call it again. Analyses consume the connected account's plan allowance; if the tool reports the allowance is used up, send the user to ViralRecreate.com/Subscription instead of retrying.
viralrecreate_get_generation_prompt
ChatGPTUse this when the user only wants one specific generation output from an existing Viral Recreate AI analysis — the image prompt, the video prompt, the movement + sound breakdown, the complete image-generation prompt, or the complete video-generation prompt — instead of the whole analysis. Read-only: it returns text already produced by the stored analysis.
