Poetry Cinema
Poetry to vertical micro-cinema films
On it. Running Poetry Cinema now and I'll bring the results straight back here.
Try asking
About
Poetry Cinema turns any poem into a fully assembled 30-second vertical short film. The pipeline takes your poem from raw text all the way to a finished micro-cinema video — scene breakdown, AI storyboard images, animated video clips, text overlays, voiceover narration, ambient sound, and final assembly with transitions and fade to black.
The production follows a four-phase workflow with review gates between each phase. Scene breakdown assigns timing, emotional beats, and colour phases to each line. Storyboard generation creates visual references for every scene. Video generation animates each storyboard into a clip. The final assembly stitches everything together with breathing text overlays, narration, and sound design.
What you can do
- Break any poem into timed scenes with emotional arcs and colour phases
- Define recurring characters with multi-angle portrait turnaround sheets for visual consistency
- Generate storyboard images for each scene in a consistent visual style
- Animate storyboard images into short video clips
- Add text overlays, voiceover narration, and ambient sound to each clip
- Assemble the final vertical film with transitions and fade to black
- Resume an existing project and pick up from where you left off
Who it's for
Poets, spoken word artists, filmmakers, content creators, and literary organisations who want to turn written work into shareable vertical video. Also useful for publishers creating book trailers, arts organisations producing digital content, and educators bringing poetry to life for students.
How to use it
- If returning to a project, use resume_project first — it reads your saved files and tells you the next step
- Use create_project to set visual style, or skip straight to break_poem with your poem text
- Review the scene breakdown, then use write_script for detailed visual direction per scene
- Progress through generate_storyboard, generate_scenes, add_layers, and assemble_film with reviews between each phase
Getting started
No API key required to get started — the tool uses the platform's default image and video models.
Check what already exists for a project and determine the next step
Set up a new poetry cinema project with production bible configuration
Define a recurring or per-poem character and generate turnaround sheets…
Divide a poem into ~12 scenes with timing, emotional arc (4-beat…
Check what already exists for a project and determine the next step. Agent reads user files via MCP tools and passes contents here. Returns what's ready, what's missing, and the next skill to call with pre-loaded inputs.
Set up a new poetry cinema project with production bible configuration. Uses scratch cinema defaults or parses an uploaded production bible document to extract visual style, colour worlds, and film rules.
Define a recurring or per-poem character and generate turnaround sheets (front/left/right/back portraits) for visual consistency across scenes. Uses the persona-generator tool for turnaround sheet creation. Takes ~70s per persona. Async — poll with get_job_result.
Divide a poem into ~12 scenes with timing, emotional arc (4-beat: hook/setup/disruption/release), and colour phase assignments. Poem text can be passed directly or read from an uploaded file. Saves scenes.json to project files.
Generate the full storyboard document from the scene breakdown — detailed visual descriptions, camera direction, sound design, and text treatment per scene. Reads scenes.json from project files. Saves script.json. This completes Phase 1 (Pre-production) — user should review before proceeding.
Generate one storyboard image per scene in the project's visual style. Includes persona references for character consistency. Use skip_scenes for user-uploaded images. Completes Phase 2 — review before proceeding.
Animate each storyboard image into a short video clip. Takes 2-5 minutes. Runs asynchronously — returns a job_id, poll with get_job_result until complete. If videos are still generating when the job completes, use check_scenes to poll for them.
Poll for pending video clips that are still generating. Pass the pending_videos array from generate_scenes output. Returns completed clips and any still-pending references.
Composite breathing text overlays, voiceover narration, and ambient sound onto each video clip. Completes Phase 3 — review layered clips before final assembly.
Stitch all layered clips into the final vertical micro-cinema film. Adds transitions (brief breath between scenes), holds final frame with last poetry line, fades to black. This completes Phase 4 (Final) — the finished film.
List available models for this tool, sorted by popularity. Returns provider details and pricing.
- Script writing now uses Gemini 3.7 Flash by default.
- Upgraded to Nano Banana 2 for storyboard images and Kling 3.0 Pro for video clips — higher quality output
- Style references now supported — pass a style reference for consistent visual direction across scenes
- Personas from your library are now auto-discovered and injected — pass persona_file_id instead of manual persona fields
- Clearer error when colour_worlds is passed as an array instead of a key-value object.
- Fix: generated scene videos now save to ToolRouter storage (were only saved to fal.ai).
- Fix: add_layers and assemble_film no longer fail with phase gate errors after running the pipeline normally.
- Fix: data flows through inputs/outputs. Better error messages. Parallel persona generation. Added resume_project for returning users.
- Initial release — poetry-to-film pipeline with scratch cinema defaults, 4-phase verification gates