Your AI can write a prompt. It doesn't know what a Kuruba woman from North Karnataka actually looks like.
Connect your AI to TASVR and it stops guessing. It searches a real catalog — 374 verified ethnicities, 5,000+ real Indian locations, wardrobe, camera, and light — and builds your image from ids that exist, not descriptions it invented.
Brief it like a creative director, not an engineer.
These are things you can actually say — your AI does the preset-finding, you keep the campaign brain.
Give me 6 Reels-ready portraits of a Gen Z model from Shillong for our streetwear drop — 9:16, punchy studio light.
Turn this one product shot into a week of Instagram carousel content — three lifestyle angles, two studio, all on-brand.
Festive campaign: a Kashmiri bride and a South Indian bride, same pose, side by side for the carousel.
We're a D2C skincare brand for urban millennial women — set up our style profile, then mock up the launch hero.
Every general model will happily guess. TASVR won't let it.
Ask for “a woman from Karnataka” and a general-purpose model renders whatever its training data associates with that phrase — often generically North Indian, often wrong. TASVR routes the same request through a real catalog instead.
A model invents.
Ask for “a woman from Karnataka” and a general model renders whatever its training data associates with that phrase — often generically North Indian, often wrong.
TASVR makes it search.
The same request has to resolve to a real preset id — Karnataka::Kuruba, Karnataka::Vokkaliga — each backed by phenotype data a human wrote and verified, not a guess an AI generated on the spot.
You can trace it back.
Every rendered detail — the face, the saree, the street — is a specific catalog entry. Ask what was used, and the answer is a real id, not “I don't know, I made it up.”
Three steps, one terminal command.
TASVR MCP is available on Pro and Enterprise plans. Get your key from Settings → Developer & API Keys first — it's shown once, so save it somewhere safe.
- 1
Create a key.
Settings → Developer & API Keys → Create key. Copy the value starting with tasvr_sk_live_.
- 2
Add the server.
Run the command from your terminal, in the project you're working in.
- 3
Confirm it connected.
Ask your AI to “list TASVR's preset categories” — a real list back means you're connected.
# adds TASVR as a remote MCP server over Streamable HTTPclaude mcp add --transport http tasvr \https://www.tasvr.ai/api/mcp \--header "Authorization: Bearer YOUR_API_KEY"
Eighteen tools across five jobs.
Your AI is expected to call the discovery tools first — it should never invent a preset id, and it can't generate anything without a real one.
The catalog's front door — every category, with a count and what it's for.
Keyword search across ~7,300 presets — “a monsoon-lit tea stall in Kerala” finds a real location, not a blank.
Every real Indian state name in the ethnicity catalog — the starting point before narrowing to a specific community.
Name a state, get back its real sub-communities with a phenotype preview each — never one silently picked for you.
Shows exactly which presets were selected — the subject, the scene, the style — before anything is spent.
Renders from chosen preset ids only. Returns instantly; poll for the result.
Polls the render. On success, the image lands in your normal Gallery — no separate MCP view.
The one deliberate exception that takes free text — for the one thing a preset list can't express.
Puts a photo you hand it into TASVR's storage — from a file or an external URL. Kept in your library only if you say so.
Keeps an uploaded reference in your library for reuse — only when you ask for it.
The 12 fixed things a reference photo can be tagged as — pose, product, jewelry, camera angle. Never a free-text caption.
Crops one specific thing out of a busy photo — a necklace, a bottle — into a clean reference. Saved to your library only if you say so.
Every avatar and reference photo you've saved in TASVR — the same library the web app itself reads from, never a separate MCP-only stash.
Remembers your tone, style, and now your ethnicity/region preference — so you're not re-asked every session.
A guided setup conversation — asks before it ever guesses who your imagery is for.
Found a gap — a missing preset, a confusing tool? This is how it reaches TASVR's team directly.
The constraints are the point.
This is what actually keeps output grounded — not a suggestion your AI can talk its way around.
generate_image never accepts a written description. Every detail in the image traces back to a real preset id your AI looked up first.
edit_image is the sole place free text is allowed — and even then it's steered by TASVR's own fixed instruction templates, not raw AI prose.
A reference photo is always labeled with one of 12 fixed roles. “Match the pose” and “ignore the pose” are both explicit — never left to a guess.
Every key has a daily spend cap, enforced atomically — a runaway agent loop can't outspend what you've set.
TASVR MCP is available on Pro and Enterprise plans. Credits, pricing, and your Gallery all work exactly as they do in the browser — MCP is a second way in, not a second account.