Make me a ~60-second, 16:9, 1080p short film from my attached photos and videos, using only the Hedr

Make me a ~60-second, 16:9, 1080p short film from my attached photos and videos, using only the Hedra MCP. Cost doesn't matter. STORY: Order the media by capture time and tell it as one day from morning to night, seen by the person holding the camera. Invent small in-between moments and about 7 short spoken lines, so it plays like a movie, not a slideshow. LOOK: For each story beat, make one 16:9 keyframe with GPT Image 2 (high). Write each prompt as a single widescreen frame from a 1990s-2000s Studio Ghibli feature film, scanned from a 35mm print: hand-painted watercolor-and-gouache backgrounds, flat cel paint with one shadow tone and thin outlines, muted slightly faded colors, soft grain. Use wide shots, with the person small in a big painted landscape. Ban: glossy, bloom, sparkles, lens flare, over-saturation, close-up portraits, text. Make the first finished keyframe the style anchor, and pass it as the second reference on every other keyframe. Show me all the keyframes as a contact sheet before animating. MOTION: Animate with Seedance 2.5 at 1080p. Clip N starts on keyframe N and ends on keyframe N+1, so the whole film is one continuous camera journey with no cuts, fades or wipes. In video prompts, describe the style instead of naming the studio (naming it trips moderation). Sound: natural ambience only (footsteps, wind, rustling, crowds). No music, and no instruments or music-like sounds that would fight the score. SCORE: One continuous 60-second live-recorded chamber waltz from ElevenLabs Music: upright piano, solo violin, accordion, mandolin, harp, small strings, flute, cello. Give it a timed arc (piano intro, playful middle, sweeping climax, tender close) ending on a soft held chord. Acoustic only, no MIDI feel, no vocals. It must never cut between scenes. Compose it in the style of Chopin. VOICE: Clone my voice from the attached recording. Voice the lines with ElevenLabs v3 at stability 0.5 and speed about 1.05, using light, playful tags such as [happy, lighthearted], [delighted], [giggles], [laughs] and [whispers], never [wistful] or [sighs]. Make 2-3 takes per line and put them on a listening page so I can pick. FINISH: Assemble with ffmpeg: 0.25 s crossfades; the score continuous under everything and ducked under the voice; scene sounds normalized per clip and sitting about 5 dB under the music; a gentle fade out. Burn in soft white film-style captions (Lato, a subtle halo, short fades) timed to the speech. Save a captioned and a clean version to my Movies folder. Check a contact sheet of stills at every caption before you call it done.