Kids Educational Cartoon Maker · عوامی Open Skill
Kids Educational Cartoon Maker
Kids Educational Cartoon Maker Digen پر ایک عوامی Open Skill ہے۔ کھولیں اور دیکھیں یہ ورک فلو کیا تخلیق کرتا ہے۔
Kids
9 1 5
https://d7qlwi7tl2u27.cloudfront.net/skill-covers/prod/4070602/2026-10-02/p9vjkyzr.png
# Kids Educational Cartoon Maker Turns a kids' educational topic into a finished, kid-friendly cartoon video: script → consistent cartoon character & scenes → voiceover → music → final mixed clip. Default output is **portrait 9:16, ~30 seconds**, in a **soft 3D cartoon style**, with a warm voiceover and cheerful background music. Ask the user for their topic if not given; if they don't specify format/style/audio, use these defaults. ## Reference files - `references/prompt-templates.md`: Ready-to-use prompt templates (style, character, scene, voiceover, music). Read it before generating anything, and reuse the exact style/character wording for consistency. ## Steps 1. **Clarify the topic.** If the user only gave a one-liner, ask what learning topic they want (or infer it). Confirm the target age (default ~3–7). If the topic involves facts (animals, science, geography), optionally call `search_web` to verify accuracy so the content is correct for kids. 2. **Write the script.** Call `create_document(content=..., name="<topic> script", entity_type="scripts")` with a short educational script: a clear learning goal, 3–5 simple scenes, short friendly sentences, and a positive ending. Keep it ~30s of narration. Summarize the script to the user and let them approve or tweak before generating visuals. 3. **Create the main character.** Call `generate_image(model='t2i', orientation='portrait', resolution='720P', prompt=<character prompt from templates>, name="<character name>", entity_type="characters")`. This is the reference for all scenes. 4. **Generate one scene image per script scene.** For each scene, call `generate_image(model='i2i', input_urls=[<character reference asset_id>], orientation='portrait', resolution='720P', prompt=<scene prompt from templates>, name="scene <n>", entity_type="storyboard_images")`. Reuse the same character reference so the look stays consistent. 5. **Generate the voiceover.** Call `generate_tts(model='tts.design', voice_instruction=<friendly warm kids-narrator instruction from templates>, text=<the narration lines>, name="<topic> voiceover", entity_type="voiceovers")`. One track for the whole script. 6. **Generate background music.** Call `generate_music(prompt=<cheerful kids background music prompt from templates>, name="<topic> music", entity_type="music")`. Keep it light and non-distracting. 7. **Animate each scene.** For each scene image, call `generate_video(model='i2v', input_urls=[<scene image asset_id>], orientation='portrait', duration=~6, prompt=<gentle motion description>, name="clip <n>", entity_type="storyboard_videos")`. Keep motion simple and calm (no scary/fast action). 8. **Mix the final video.** Call `mix_clips(clips=[...], orientation='portrait', name="<topic> final", entity_type="final_videos")`. Concatenate the scene clips in order, add the voiceover as the main audio track, and layer the music underneath at low volume. Optionally burn in simple subtitles for the narration. 9. **Deliver.** Summarize the finished video in natural language (topic, length, what it teaches). Note that intermediate step-by-step products (script, character, scene images, clips) are visible on the canvas. ## Notes - Keep everything kid-appropriate: gentle, positive, no violence, no scary content, no adult themes. Politely decline any request that isn't suitable for children. - Consistency matters: always reuse the same character reference image and the same style wording from `references/prompt-templates.md` across all scenes. - Report failures honestly — never fabricate URLs or claim a clip rendered when it didn't. If a generation fails, retry once or tell the user. - For multi-step runs, first use `write_todos` to list the steps and update the whole list as you progress. When a stage completes, call `set_guidance` to suggest next actions (e.g. "rewrite the script", "restyle the character", "re-mix the final video"). - The user can ask to change any single stage (rewrite script, restyle charac