Create an authentic 30-second smartphone vlog from the user's text and optional REFERENCE image. Show one connected real-life experience, not a stretched clip, slideshow, trailer or advertisement. PRIORITY Text controls topic, actions, location, viewpoint, mood and speech. A reference image controls WHO/WHAT appears, not the opening frame, crop, pose or background if text specifies another scene. Explicit requests outrank defaults. Without an image, invent suitable subjects and setting. Do not let an indoor portrait override an outdoor request. SMART COMPLETION Handle both detailed descriptions and just a few words. Internally choose a main subject, one plausible activity, suitable setting, connected story beats, camera style and ambience. Topic only: select a specific believable experience. Mood/style only: infer a simple compatible activity from the image; without a photo, choose an ordinary activity. Generic "vlog": create one natural everyday moment. Add missing essentials, but not named destinations, personal histories, relationships, brands, unrelated people or dramatic plots without a basis. Respect every specific request. PHOTO ADAPTATION Headshot: preserve face, skin, hair, visible clothes/accessories; conservatively complete unseen body/clothing for wider shots. Half/full-body: preserve recognizable build and outfit. Selfie: identity only, unless selfie filming is requested. Group: retain intended people, distinct identities and count. Pet: lock species, markings, size and anatomy. Product: lock quantity, proportions, design, material, colors, logos and readable labels. Food/landscape: preserve distinctive features unless a change is requested. If user specifies a new place, replace the source background while retaining the subject. Never force a presenter into pure scenery; add hands only where the activity needs them. STORY: 30 SECONDS 0-5s: immediately establish the specified place, subject and activity. 5-12s: first meaningful interaction or discovery. 12-23s: 1-2 connected developments showing fresh details and actual progress. 23-30s: a relevant outcome, discovery or relaxed response; end on a living frame. These are flexible beats, NOT mandatory cuts. Choose an achievable 30-second experience, not a whole day of disconnected highlights. Avoid looping movements, repeated smiles, static portraits and filler. CAMERA & EDITING Normally use 3-5 purposeful phone-camera shots; fewer for a natural one-take. Include an establishing medium/wide