Start an async AI video generation job — returns a job_id immediately.
Start an async AI video generation job — returns a job_id immediately. Poll with check_status every 10-30s until complete. Cost varies by model: veo3-fast (~15 credits/5s), kling-3 (~30 credits/5s), sora2-pro (~60 credits/10s). Check get_credit_balance first for expensive generations. Costs ~5-50 credits.
Start an async AI video generation job — returns a job_id immediately. › Request Body
promptVideo prompt — be specific about visual style, camera movement, lighting, and mood. Example: "Aerial drone shot of coastal cliffs at golden hour, slow dolly forward, cinematic 24fps, warm color grading." Vague prompts produce generic results.
modelVideo model. veo3-fast: fastest (~15 credits/5s, ~60s render). veo3-quality: highest quality (~20 credits/5s, ~120s). sora2-pro: OpenAI premium (~60 credits/10s). kling-3: 4K with audio (~30 credits/5s). kling-3-pro: best Kling quality (~40 credits/5s).
durationVideo duration in seconds. kling: 5-30s, kling-3/kling-3-pro: 3-15s, ' + 'sora2: 10-15s. Defaults to 5 seconds.
aspect_ratioVideo aspect ratio. 16:9 for YouTube/landscape, 9:16 for TikTok/Reels/Shorts, 1:1 for Instagram feed/square. Defaults to 16:9.
enable_audioEnable native audio generation. Kling 2.6: doubles cost. ' + 'Kling 3.0: 50% more (std 30/sec, pro 40/sec). 5+ languages.
image_urlStart frame image URL for image-to-video (Kling 3.0 frame control).
end_frame_urlEnd frame image URL (Kling 3.0 only). Enables seamless loop transitions.
response_formatOptional response format. Defaults to text.
Start an async AI video generation job — returns a job_id immediately. › Responses
Successful tool execution.
Start an async AI image generation job — returns a job_id immediately.
Start an async AI image generation job — returns a job_id immediately. Poll with check_status every 5-15s until complete. Costs 2-10 credits depending on model. Use for social media posts, carousel slides, or as input to generate_video (image-to-video). Costs ~5-50 credits.
Start an async AI image generation job — returns a job_id immediately. › Request Body
promptText prompt describing the image to generate. Be specific about style, ' + 'composition, colors, lighting, and subject matter.
modelImage generation model. midjourney for artistic style, imagen4 for ' + 'photorealistic quality, flux-pro for general purpose, gpt4o-image ' + 'for creative/illustrated styles.
aspect_ratioAspect ratio. Defaults to 1:1 (square).
image_urlReference image URL for image-to-image generation. Required for ' + 'ideogram model. Optional for others.
response_formatOptional response format. Defaults to text.
Start an async AI image generation job — returns a job_id immediately. › Responses
Successful tool execution.
Poll an async job started by generate_video or generate_image.
Poll an async job started by generate_video or generate_image. Returns status (queued/processing/completed/failed), progress %, and result URL on completion. Poll every 10-30s for video, 5-15s for images. On failed status, the error field explains why — check credits or try a different model. Costs ~5-50 credits.
Poll an async job started by generate_video or generate_image. › Request Body
job_idThe job ID returned by generate_video or generate_image. ' + 'This is the asyncJobId or taskId value.
response_formatOptional response format. Defaults to text.
Poll an async job started by generate_video or generate_image. › Responses
Successful tool execution.
Plan a multi-scene video storyboard with AI-generated prompts, durations, captions, and voiceover text per frame.
Plan a multi-scene video storyboard with AI-generated prompts, durations, captions, and voiceover text per frame. Use before generate_video or generate_image to create cohesive multi-shot content. Include brand_context from get_brand_profile for consistent visual branding across frames. Costs ~5-50 credits.
Plan a multi-scene video storyboard with AI-generated prompts, durations, captions, and voiceover text per frame. › Request Body
conceptThe video concept/idea. Include: hook, key messages, target audience, ' + 'and desired outcome (e.g., "TikTok ad for VPN app targeting ' + 'privacy-conscious millennials, hook with shocking stat about data leaks").
platformTarget platform. Determines aspect ratio, duration, and pacing.
brand_contextBrand context JSON from extract_brand. Include colors, voice tone, ' + 'visual style keywords for consistent branding across frames.
target_durationTarget total duration in seconds. Defaults to 30s for short-form, 60s for YouTube.
num_scenesNumber of scenes. Defaults to 6-8 for short-form.
styleVisual style direction (e.g., "cinematic", "anime", "documentary", "motion graphics").
response_formatResponse format. Defaults to json for structured storyboard data.
Plan a multi-scene video storyboard with AI-generated prompts, durations, captions, and voiceover text per frame. › Responses
Successful tool execution.
Generate a voiceover audio file for video narration.
Generate a voiceover audio file for video narration. Returns an R2-hosted audio URL. Use after create_storyboard to add narration to each scene, or standalone for podcast intros and ad reads. Costs ~2 credits per generation. Costs ~5-50 credits.
Generate a voiceover audio file for video narration. › Request Body
textThe script/text to convert to speech.
voiceVoice selection. rachel=warm female, drew=confident male, ' + 'paul=authoritative male, sarah=friendly female. Defaults to rachel.
speedSpeech speed multiplier. 1.0 is normal. Defaults to 1.0.
response_formatResponse format. Defaults to text.
Generate a voiceover audio file for video narration. › Responses
Successful tool execution.
Generate carousel slide content (headlines, body text, emphasis words per slide).
Generate carousel slide content (headlines, body text, emphasis words per slide). Supports Hormozi-style authority format and educational templates. Returns structured slide data — render visually then publish via schedule_post with media_type=CAROUSEL_ALBUM and 2-10 media_urls on Instagram. Costs ~5-50 credits.
Generate carousel slide content (headlines, body text, emphasis words per slide). › Request Body
topicCarousel hook/angle — specific beats general. Example: "5 pricing mistakes that kill SaaS startups" beats "SaaS tips". Include a curiosity gap or strong opinion for better Hook Strength scores.
template_idCarousel template. hormozi-authority: bold typography, one idea per slide, ' + 'dark backgrounds. educational-series: numbered tips. Default: hormozi-authority.
slide_countNumber of slides (3-10). Default: 7.
aspect_ratioAspect ratio. 1:1 square (default), 4:5 portrait, 9:16 story.
styleVisual style. hormozi: black bg, bold white text, gold accents. ' + 'Default: hormozi (when using hormozi-authority template).
hookExplicit hook/opener for slide 1. Overrides any hook derived from topic. Keep under 15 words.
hook_familyHook family tag. Persisted with the carousel so bandit learners can attribute engagement to hook pattern.
cta_textExplicit CTA copy for the final slide.
cta_urlURL promoted on the CTA slide.
toneVoice/tone override. Composes with brand profile voice.
constraintsContent constraints. Example: "No fabricated statistics. Sentence case only."
platformTarget platform. Affects tone and format guardrails.
project_idProject ID to associate the carousel with.
response_formatResponse format. Defaults to json.
Generate carousel slide content (headlines, body text, emphasis words per slide). › Responses
Successful tool execution.