See the shot
Subjects, scene, composition, palette, lighting, texture, and visible action.
Built for AI video creators
Video to Prompt generator
for shots worth recreating.
Upload a short reference video. Get its camera language, motion, lighting, timing, audio—and production-ready prompts for the models you use.
One analysis, formatted forUniversalVeo 3KlingSeedance
Shot analyzer
Upload a reference clip
Uploaded video file not retained
Drop your video here
or click to browse your files
MP4, MOV, WebM Up to 15 seconds Beta max 4 MB
Real example output
See the shot become usable direction.
This original five-second reference clip was analyzed by ShotToPrompt. The excerpts below are from its generated result, not placeholder copy.
Original ShotToPrompt demo · 5 seconds
01 · Visual DNA
Minimalist cyclist in a rainy night city
Locked side-profile tracking, low-key nocturnal light, a dark navy palette, warm orange subject, and continuous horizontal motion.
02 · Timeline
00:00 → 00:04
The rider pedals continuously while the camera follows at a fixed speed; soft rain ambience and subtle mechanical sounds support the motion.
03 · Universal prompt excerpt
“A minimalist 2D vector animation of an orange figure riding a white line-art bicycle across a dark street at night… The camera follows from a strict side-profile perspective in a smooth horizontal tracking shot.”
From reference to direction
More than “describe this video.”
ShotToPrompt translates what the clip does into the production choices an AI video model can follow.
Read the movement
Lens feel, camera position, motion path, transitions, pace, and temporal beats.
Direct the model
Editable prompts tailored to the strengths and conventions of each video model.
How the analysis works
Your clip is analyzed, not warehoused.
Your video is sent directly to Google Gemini for visual and audio analysis. ShotToPrompt does not save the uploaded video itself.
We keep the generated prompt, basic file metadata, and an anonymous usage identifier for a limited period so the tool can enforce its daily limit and improve reliability.
See exactly what we process and retainClear before you upload
Frequently asked
Does this recover the original prompt?
No. A video does not contain its source prompt. ShotToPrompt analyzes observable production choices and creates a new, editable prompt that aims to reproduce its visual grammar.
What video formats work?
MP4, MOV, WebM, and MPEG clips up to 15 seconds. The temporary beta upload limit is 4 MB. Short, visually coherent clips produce the most useful shot breakdowns.
Do you keep my video?
No. The video is sent to Gemini for the analysis request and is not saved in the ShotToPrompt database. We retain the generated analysis and minimal operational metadata for limits and product quality.
Why different prompts for each model?
Video models respond differently to motion, timing, audio, and structure. Each version preserves the same creative intent while emphasizing the controls that model handles best.