MCP methods reference
Everything your assistant can create for you
51 tools across 9 groups: generation, editing, characters, films, templates, your library and brand. Each one works from a normal chat message. No code, no copy-paste. This page is what your assistant sees.
28 are live on the MCP endpoint today; the rest ship phase by phase.
01 — Create
Make something from nothing
Ask for images, video and voice from the top models on the market. One prompt, one answer, or several to compare.
generate_imageHeadline LiveGenerate an image
Create a new image from a text prompt, or edit one by passing image URLs. Works with Nano Banana, GPT Image, FLUX, Seedream, Recraft and more.
Say to your assistant
“Make a product shot of a matte black espresso machine on a warm kitchen counter, golden hour.”
Say to your assistant
“Put this logo on a coffee cup, photorealistic.”
From 6 credits
Options (7)
- prompt
- What to make, in plain words
- model
- Optional; the recommended default is used otherwise
- image_urls
- Reference or source images for edits
- aspect_ratio
- 1:1, 4:5, 9:16, 16:9 and more
- num_images
- How many variants, 1–4
- theme_id
- A saved brand theme to follow
- character_id
- A saved Character to keep consistent
generate_videoHeadline LiveGenerate a video
Produce video from text, from a single start frame, or from first and last frames. Draft at 480p first, then render the one you like.
Say to your assistant
“Make a 6-second clip of a runner sprinting at dawn, 9:16.”
Say to your assistant
“Animate this photo with a slow camera push-in.”
From 5 credits a second
Options (7)
- prompt
- The shot, the motion and the sound
- model
- Seedance, Kling, Gemini Omni Flash, MiniMax and more
- image_url
- A start frame to animate
- end_image_url
- An optional last frame to land on
- duration
- Seconds, model limits apply
- resolution
- 480p drafts up to 4K
- aspect_ratio
- 16:9 or 9:16
generate_speechHeadline LiveGenerate speech
Turn any text into studio-quality narration with a voice, a language and a delivery. Use a stock voice, a designed one or a saved Character’s.
Say to your assistant
“Read this script in a warm, confident female voice.”
Say to your assistant
“Do the same line in Slovenian.”
20 credits per 1,000 characters
Options (5)
- text
- The script; supports tags like [whispering]
- voice
- A stock, designed or cloned voice
- language
- ISO language code
- stability
- Steadier or more expressive delivery
- character_id
- Speak as a saved Character
generate_musicLiveGenerate music
Background music, jingles or full songs. Lyria for quick beds, ElevenLabs Music for an exact length, MiniMax or Mureka to sing your lyrics.
Say to your assistant
“A 30-second upbeat lo-fi bed for this reel.”
Say to your assistant
“A short jingle that sings our brand name.”
20 credits a track; 120 a minute for an exact length
Options (4)
- prompt
- Genre, mood, instruments
- duration
- Length where the model supports it
- lyrics
- Words to sing
- instrumental
- No vocals
generate_sound_effectLiveGenerate a sound effect
Sound effects, Foley, ambience and UI sounds from a description, 0.5 to 30 seconds, with seamless loops for ambience.
Say to your assistant
“A heavy door slamming in a stone hallway.”
Say to your assistant
“Ten seconds of rain on a tent, looping.”
About 0.4 credits a second
Options (3)
- prompt
- What it sounds like
- duration
- 0.5–30 seconds
- loop
- Seamless loop
recommend_video_modelLivePick the right video model
Reads your brief (length, shape, sound, speech, start image, style, budget) and returns the best three models with exact credits and trade-offs.
Say to your assistant
“Which model should I use for a 10-second product ad with voice?”
Say to your assistant
“What’s the cheapest way to draft this?”
Free
Options (4)
- prompt
- What you want to make
- duration
- Length
- priority
- Cheapest, balanced or best quality
- budget_credits
- A credit ceiling
compare_modelsAt launchCompare models side by side
Run the same brief on two to four models at once and get the results back together, with each one’s cost, so you can pick the winner.
Say to your assistant
“Try this poster on GPT Image and Nano Banana and show me both.”
Say to your assistant
“Which video model handles this motion best? Test three at 480p.”
Sum of the models used
Options (3)
- prompt
- The shared brief
- models
- Two to four model ids
- image_urls
- Shared references
02 — Edit & transform
Polish what you have
Upscale, cut out, trim, reframe, subtitle and transcribe. Each transform is its own tool, so your assistant can chain them.
upscale_imageLiveUpscale an image
Enlarge an image up to 4× with recovered detail, ready for print or a hero banner.
Say to your assistant
“Upscale the winner to 4K.”
Say to your assistant
“Make this logo sharp enough for a billboard.”
2 or 80 credits
Options (2)
- image_url
- The image to enlarge
- scale
- 2× or 4×
remove_backgroundLiveRemove a background
Cut the subject out of an image or a video and return it on transparency, ready for a new scene.
Say to your assistant
“Remove the background from all 12 product photos.”
Say to your assistant
“Give me this shoe on transparent.”
2 credits an image
Options (2)
- media_url
- Image or video
- output
- Transparent PNG, or a solid colour
upscale_videoVideoUpscale a video
Raise a clip’s resolution with temporal consistency, so a 480p draft becomes a 1080p or 4K master.
Say to your assistant
“Upscale the draft to 1080p.”
Say to your assistant
“Make this old clip look sharp.”
Measured per video
Options (2)
- video_url
- The clip
- target
- 1080p or 4K
increase_frame_rateVideoIncrease the frame rate
Interpolate new frames for smoother motion or buttery slow motion.
Say to your assistant
“Make this clip 60 fps.”
Say to your assistant
“Slow the splash down to quarter speed, smoothly.”
About 5 credits a second
Options (3)
- video_url
- The clip
- fps
- Target frames per second
- slow_motion
- Speed factor
trim_videoLiveTrim a video
Cut a clip to a start and end time, or to several segments in order.
Say to your assistant
“Keep 0:03 to 0:09.”
Say to your assistant
“Cut it down to the best 8 seconds.”
2 credits
Options (3)
- video_url
- The clip
- start
- Seconds or timestamp
- end
- Seconds or timestamp
reframe_videoLiveReframe or resize a video
Change the shape, for example 16:9 to 9:16 for TikTok and Reels, by cropping to fill the frame or fitting it with bars.
Say to your assistant
“Make a 9:16 version for TikTok.”
Say to your assistant
“Square crop for the feed.”
2 credits
Options (3)
- video
- The clip
- aspect_ratio
- Target ratio
- fit
- Crop to fill, or pad with bars
extract_framesVideoExtract frames from a video
Pull stills at timestamps or at an interval, for thumbnails, storyboards or the next start frame.
Say to your assistant
“Grab a frame every second.”
Say to your assistant
“Give me the last frame as a start image.”
2 credits
Options (3)
- video_url
- The clip
- times
- Timestamps, or
- every
- An interval in seconds
transcribeAudioTranscribe audio or video
Get a word-timed transcript in dozens of languages, with speakers labelled.
Say to your assistant
“Transcribe this interview.”
Say to your assistant
“What does she say at the 2-minute mark?”
About 96 credits an hour
Options (3)
- media_url
- Audio or video
- language
- Optional; detected otherwise
- speakers
- Label speakers
isolate_audioAudioClean a voice recording
Remove background noise, music and room echo, and keep the voice.
Say to your assistant
“Clean up the street noise in my voice note.”
Say to your assistant
“Separate the voice from the music.”
80 credits
Options (2)
- audio_url
- Audio or video
- keep
- Voice, music or both as stems
add_captionsLiveBurn in captions
Burn word-by-word captions into a video, TikTok style. Uses the timings from a voiceover, or transcribes the speech itself.
Say to your assistant
“Add bold captions, one word at a time.”
Say to your assistant
“Put clean subtitles on this interview.”
6 credits a minute, plus ~2 to transcribe
Options (3)
- video
- The clip
- words
- Word timings from generate_speech (optional)
- style
- Bold pop-up words or clean lines
03 — Combine
Stitch media together
Put a voiceover on a clip, join clips in order or mix tracks into one deliverable.
merge_mediaLiveMerge audio, video or both
Concatenate clips, lay audio on video, or mix music under a voiceover with ducking.
Say to your assistant
“Put the voiceover on the clip with the music quietly underneath.”
Say to your assistant
“Join these three shots in order.”
2+ credits
Options (4)
- video_urls
- Clips, in order
- audio_urls
- Voice, music or effects
- mix
- Volumes and ducking
- transitions
- Cut, fade or crossfade
04 — Templates & layout
Branded graphics from HTML
Render exact layouts from HTML and CSS, or save templates with placeholders. Faster, cheaper and pixel-perfect when AI is overkill.
render_html_imageTemplatesRender HTML to an image
Turn HTML and CSS into a PNG at any size, with your fonts and images.
Say to your assistant
“Make an OG image with this headline and my logo.”
Say to your assistant
“Render this pricing table as a PNG.”
1 credit
Options (4)
- html
- The markup
- width
- Pixels
- height
- Pixels
- inline_images
- Images to embed
create_templateTemplatesSave a reusable template
Store a layout with {{token}} placeholders, so the same design renders with new content every time.
Say to your assistant
“Save this as my quote-card template.”
Say to your assistant
“Make a template for weekly sale banners.”
Free
Options (3)
- name
- Template name
- html
- Markup with {{tokens}}
- static_assets
- Fonts, logos, backgrounds
render_templateTemplatesRender a template
Fill a template’s tokens and render one image or hundreds in a single call.
Say to your assistant
“Render the sale banner for these 50 products.”
Say to your assistant
“Make this week’s quote card.”
4 credits a render
Options (2)
- template_id
- Which template
- values
- Token values, one set per image
update_templateTemplatesUpdate a template
Change a saved template’s layout, assets or tokens without starting over.
Say to your assistant
“Make the headline bigger on the banner template.”
Say to your assistant
“Swap the background image.”
Free
Options (3)
- template_id
- Which template
- html
- New markup
- static_assets
- New assets
list_templatesTemplatesList your templates
See every saved template with its tokens and a preview.
Say to your assistant
“What templates do I have?”
Say to your assistant
“Find my banner templates.”
Free
Options (1)
- name_contains
- Filter by name
05 — Characters
Reuse the same person everywhere
Save a face, outfit and voice once. Pass the Character into images, video and speech so identity and delivery stay the same.
manage_characterCharacters & filmsCreate or update a Character
Build a reusable Character from photos or a description, with a reference sheet and an optional voice.
Say to your assistant
“Make a Character from these three photos and call her Nora.”
Say to your assistant
“Give Nora a warm, friendly voice.”
From 16 credits
Options (4)
- name
- Character name
- image_urls
- Reference photos
- description
- Look and personality
- voice
- Designed, cloned or stock
list_charactersCharacters & filmsList your Characters
See every saved Character with their reference sheet and voice.
Say to your assistant
“Who are my Characters?”
Say to your assistant
“Show me Nora’s reference sheet.”
Free
Options (1)
- name_contains
- Filter by name
delete_characterCharacters & filmsDelete a Character
Remove a Character and its cloned voice. Media already made with it stays in your library.
Say to your assistant
“Delete the test Character.”
Say to your assistant
“Remove my old avatar.”
Free
Options (1)
- character_id
- Which Character
swap_video_charactersLiveSwap characters in a video
Replace a person in an existing video with a saved Character, keeping the performance and camera move.
Say to your assistant
“Put Nora into this dance clip.”
Say to your assistant
“Swap me in, preview a frame first.”
From 10 credits a second
Options (3)
- video_url
- The source video
- character_id
- Who to put in
- preview
- Return a still first
06 — Films
Build multi-shot films with approval gates
Plan a story, approve the script and storyboard, generate connected shots, then assemble the cut.
create_film_projectCharacters & filmsCreate a film project
Start a film with a story, a target length, a format and the Characters in it.
Say to your assistant
“Plan a 30-second film about a runner’s first marathon.”
Say to your assistant
“Make it 9:16 with Nora as the lead.”
Free
Options (4)
- title
- Working title
- story
- The idea
- length_seconds
- Target length
- character_ids
- The cast
update_film_projectCharacters & filmsUpdate a film project
Save the script, the shot list, storyboards and approvals as the film moves along.
Say to your assistant
“Approve the storyboard but redo shot 3.”
Say to your assistant
“Make the ending happier.”
Free
Options (4)
- film_id
- Which film
- script
- Script text
- shots
- Shot list and frames
- status
- Approval stage
get_film_projectCharacters & filmsGet a film project
Read a film’s current script, shots, clips and what is waiting for approval.
Say to your assistant
“Where are we on the marathon film?”
Say to your assistant
“Show me the shot list.”
Free
Options (1)
- film_id
- Which film
list_film_projectsCharacters & filmsList film projects
See every film with its stage, length and last update.
Say to your assistant
“What films am I working on?”
Say to your assistant
“Open the latest one.”
Free
Options (1)
- status
- Filter by stage
assemble_filmCharacters & filmsAssemble the final film
Cut the approved clips together with narration, music, captions and an end card.
Say to your assistant
“Assemble the cut with narration and a music bed.”
Say to your assistant
“Add captions and my logo at the end.”
From 4 credits
Options (4)
- film_id
- Which film
- include_narration
- Lay in the voiceover
- music
- A track or a mood
- captions
- Burn in subtitles
07 — Your library
Organise every asset you create
Every result is stored with a permanent link. Search it, tag it, upload your own files and reuse anything across chats.
search_assetsLiveSearch your library
Find anything you’ve made or uploaded by words, type, tag or date.
Say to your assistant
“Find the KAIRO ads from last week.”
Say to your assistant
“Show me every video I made in 9:16.”
Free
Options (3)
- query
- What to look for
- type
- Image, video or audio
- tags
- Tags to match
upload_assetAt launchUpload a file
Save an image, video or audio file to your library from a URL or the chat.
Say to your assistant
“Save this photo to my library.”
Say to your assistant
“Keep this logo for later.”
Free
Options (3)
- url
- Where the file is
- name
- A friendly name
- tags
- Tags
get_upload_urlLiveLarge file upload (step 1)
Get a private upload link for files too big to send through the chat.
Say to your assistant
“I want to upload a 2 GB interview.”
Say to your assistant
“Give me a link to upload my footage.”
Free
Options (2)
- filename
- The file’s name
- content_type
- Its media type
confirm_uploadLiveLarge file upload (step 2)
Tell Typeshoot the upload finished, so the file appears in your library.
Say to your assistant
“The upload is done.”
Say to your assistant
“Check my footage arrived.”
Free
Options (1)
- asset_id
- From step 1
analyze_videoLiveAnalyze a video
Breaks a video into shots with timestamps, shows the key frames and transcribes the speech, so the assistant can explain the hook, structure and pacing before you recreate or swap it.
Say to your assistant
“Why does this TikTok work? Break it down.”
Say to your assistant
“Analyze this ad and make one like it for my product.”
About 2 credits a minute
Options (2)
- media_id
- A video from import_media
- url
- Or a direct link
import_mediaLiveImport from a link
Bring in media from a public video or social link, ready to edit or transcribe.
Say to your assistant
“Import this YouTube video.”
Say to your assistant
“Grab the clip from this post.”
1 credit
Options (2)
- url
- A public link
- format
- Video, audio only or both
load_imageFoundationLook at an image
Let your assistant see an image from your library or a URL, so it can describe, judge or compare it.
Say to your assistant
“Which of the four ads reads best at phone size?”
Say to your assistant
“Describe what’s in this photo.”
Free
Options (1)
- url
- Image URL or asset id
update_assetAt launchRename or tag an asset
Change an asset’s name, description or tags.
Say to your assistant
“Tag these as KAIRO launch.”
Say to your assistant
“Rename this to final hero.”
Free
Options (3)
- asset_id
- Which asset
- name
- New name
- tags
- New tags
delete_assetAt launchDelete an asset
Remove an asset from your library. Deletes are soft for 30 days, so mistakes can be undone.
Say to your assistant
“Delete the rejected drafts.”
Say to your assistant
“Remove this image.”
Free
Options (1)
- asset_id
- Which asset
08 — Brand themes
Stay on brand automatically
Save colours, fonts, logos and voice guidelines once. Every generation can use your theme, so you never paste a brand kit again.
update_themeLiveSave a brand theme
Create or change a theme from uploads, your website or a description.
Say to your assistant
“Save my brand from kairo.com.”
Say to your assistant
“Add volt lime as an accent colour.”
Free
Options (6)
- name
- Theme name
- logo_url
- Logo
- colors
- Hex colours
- fonts
- Type
- tone
- Voice guidelines
- set_default
- Use it by default
get_themeLiveGet a brand theme
Read a theme, so your assistant knows the rules before it writes a prompt.
Say to your assistant
“What’s in my brand kit?”
Say to your assistant
“Use my theme for this.”
Free
Options (1)
- theme_id
- Optional; the default otherwise
list_themesLiveList brand themes
See every theme, handy when you work for more than one brand.
Say to your assistant
“Which brands have I saved?”
Say to your assistant
“Switch to the client theme.”
Free
delete_themeLiveDelete a brand theme
Remove a theme you no longer need.
Say to your assistant
“Delete the old brand.”
Say to your assistant
“Remove the test theme.”
Free
Options (1)
- theme_id
- Which theme
09 — Discovery & account
Pick models, track jobs, manage credits
The layer underneath. Most of it runs on its own, but you can call any of it when you want control.
list_modelsLiveList models
Browse the models for a task, with what each is good at and its price.
Say to your assistant
“Which video models can do native audio?”
Say to your assistant
“What’s the cheapest good image model?”
Free
Options (2)
- category
- Image, video, voice, music…
- query
- A capability to look for
get_model_paramsLiveGet a model’s options
Read one model’s inputs, limits and defaults before using it.
Say to your assistant
“What can Seedance 2.5 do?”
Say to your assistant
“How long can a Kling clip be?”
Free
Options (1)
- model
- The model id
estimate_generationLiveEstimate the cost
Price a generation before it runs. Anything over 100 credits asks first.
Say to your assistant
“How much would a 10-second 1080p clip cost?”
Say to your assistant
“Estimate the whole campaign.”
Free
Options (2)
- model
- The model id
- params
- The settings you would use
check_jobLiveCheck a generation job
Get a job’s status, and its permanent file link once it’s done.
Say to your assistant
“Is the video ready?”
Say to your assistant
“Show me the result.”
Free
Options (1)
- job_id
- From the generate call
get_credits_balanceLiveCheck your credit balance
See your balance and what you’ve spent recently.
Say to your assistant
“How many credits do I have?”
Say to your assistant
“What did yesterday cost?”
Free
get_credits_linkLiveBuy more credits
Get a checkout link for a credit pack. Credits never expire.
Say to your assistant
“Top me up with $10.”
Say to your assistant
“Send me the link to buy credits.”
Free
Options (1)
- pack
- Which credit pack
Connect it in under a minute
One MCP URL. Every tool on this page becomes available in your assistant the moment you connect and sign in.
https://typeshoot.ai/mcpClaude
ChatGPT
Claude Code
Codex
Cursor