Install Jentic One Beta
Jentic One is a self-hosted execution layer for AI agents. It lets your agent call the MiroMiro Design Extraction API, or any other public or private API you need. You set the rules, the agent never sees your credentials, and every call is logged.
Two steps, two machines. Install the instance in a safe environment, then register your agent from wherever it runs.
Step 1: Jentic One Host machine
# On the machine that will host your Jentic One instance:
curl -fsSL "https://jentic.com/install.sh?src=apis&api=%2Fapis%2Fmiromiro.app%2Fmiromiro" | shStep 2: Agent machine
# On the machine where your agent runs (keep this separate from the instance):
curl -fsSL "https://jentic.com/install.sh?src=apis&api=%2Fapis%2Fmiromiro.app%2Fmiromiro" | sh
jentic register # connects your agent to your Jentic One instanceJentic One is in public beta. The setup above keeps your agent separate from the instance, which is what you want before using real credentials: an agent running as the same OS user as Jentic One can read its stored keys directly. Just evaluating? A single local install is fine to start. See the secure deployment guide for the tiers.
What an agent can do with MiroMiro Design Extraction API.
Extract a page section such as hero, pricing, or footer as a Tailwind, HTML, JSX, or Vue component
Return a site's brand identity: color roles, palette, primary fonts, logos, og-image, and social profiles
Export the full design-token system as JSON, CSS custom properties, a Tailwind theme, or DTCG
List every @font-face file a site loads with family, weight, style, and Google Fonts links
GET STARTED
Collect a page's images, inline and external SVGs, videos, audio files, and Lottie animations
Render client-side JavaScript sites in a real browser and capture a viewport or full-page screenshot
Patterns agents use MiroMiro Design Extraction API for, with concrete tasks.
★ Agent UI Building Through Jentic
A coding agent asked to build a page in the style of a reference site pulls the hero or pricing section from that URL as a ready-made Tailwind or React component, along with the tokens it uses, then adapts it in your codebase. Through Jentic the agent finds the code operation by intent and calls it with url, section, and framework, using the stored key without seeing it.
Extract the pricing section of 'stripe.com' with framework 'jsx' and name 'PricingTable', and return the component code
Brand Kits from a Domain
Sales, onboarding, and white-label tools personalise their UI by pulling a customer's brand from their website: named primary, accent, background, and text colors, the palette, fonts by role, logos, and social links. The confidence score shows how much the site declared versus what was inferred, which is the value to gate on for bulk runs.
Get the brand for 'linear.app' and return the primary and accent colors, the heading font, and the confidence score
Design-Token Audits and Migrations
Design-system teams export a site's tokens, including colors, font sizes, spacing, border radii, shadows, gradients, and breakpoints with usage counts, as DTCG or CSS custom properties to compare against their own system or seed a migration. Font and SVG extraction fill in the assets the tokens refer to.
Extract design tokens from 'vercel.com' in format 'dtcg' and list the fonts the site loads
9 endpoints — the miromiro design extraction api turns a live website url into design data you can build with.
METHOD
PATH
DESCRIPTION
/code
Extract a page section as a component
/brand
Brand identity: colors, fonts, logo, socials
/extract
The full design-token system of a page
/fonts
The typography a site loads
/svg
Inline and external SVGs
/images
Every image a page declares
/code
Extract a page section as a component
/brand
Brand identity: colors, fonts, logo, socials
/extract
The full design-token system of a page
/fonts
The typography a site loads
/svg
Inline and external SVGs
/images
Every image a page declares
What agents get from Jentic-routed access to this vendor.
Setup
Wiring MiroMiro by hand means setting the Bearer header, deciding per call when render=true is worth its extra credits, and trimming large responses with the fields parameter. Through Jentic you install once, import MiroMiro from the API Directory, store the key once, and your agent calls it.
Permission scoping
MiroMiro takes the target page as a url query parameter on fixed paths, so rules bound which operations your agent may call, not which sites it extracts from. You choose the operations it may call, so credit-heavy ones like code or brand extraction are not included unless you add them.
Credential isolation
Your MiroMiro key is stored once, encrypted, by your own Jentic One instance and injected at execution time. It never enters the agent's prompt, logs, or context.
Intent-based discovery
Agents search Jentic by intent such as 'turn a website section into code' or 'get a site's brand colors', and Jentic returns the matching MiroMiro operation with its input schema so the agent calls the right endpoint without reading the reference docs.
Alternatives and complements available in the Jentic catalogue.
Specific to using MiroMiro Design Extraction API through Jentic.
What authentication does the MiroMiro Design Extraction API use?
The MiroMiro Design Extraction API authenticates with a key sent as `Authorization: Bearer mm_live_...` per its OpenAPI spec, declared globally for every operation. Through Jentic the key is stored once, encrypted, by your own Jentic One instance and injected at call time, so it never enters the agent's prompt or logs.
Which frameworks can the MiroMiro code operation output?
The framework parameter accepts tailwind (the default), html, jsx, or vue. Pick a section by name such as hero, header, footer, pricing, features, testimonials, faq, or cta, or pass a CSS selector; omit both and the hero is auto-detected, with the response listing every other extractable section.
Can I limit what my agent is allowed to do with the MiroMiro API?
Yes. For example, write a rule that allows only the brand and fonts operations, so the agent can pull brand colors and typography but not run the higher-cost code extraction, and every call it makes is logged.
What are the rate limits for the MiroMiro Design Extraction API?
The OpenAPI spec meters usage in credits rather than stating a request rate: for example brand costs 15 credits, fonts 5, images and SVGs 1 each, and render=true adds 5, with cache hits and failed calls free. A free key includes 300 credits a month; see https://miromiro.app/api/docs for current limits.
How do I extract a website section with MiroMiro through Jentic?
Search Jentic for the intent 'turn a website section into code', which returns the MiroMiro code operation on /code with its input schema; the agent passes url, section or selector, and framework, adding render=true for client-rendered sites. To run it on your own infrastructure, install Jentic One from its GitHub repo.
Is there a MiroMiro MCP server?
You don't need an MCP server to give your agent the MiroMiro API. Jentic connects it directly from the API Directory: import it, store your key once, and your agent calls the code, brand, token, and asset operations. Operations are discovered on demand, so no extra tool definitions are loaded into the agent's context.
For Agents
Extract a live website's sections as Tailwind, HTML, React, or Vue components, plus its brand identity, design tokens, fonts, images, SVGs, videos, audio, and Lottie animations with MiroMiro. Authenticated with a Bearer key.
Use for: Turn the pricing section of this site into a React component, Get the brand colors and fonts for stripe.com, Export this site's design tokens as a Tailwind theme, Find all the fonts a website loads
Not supported: Does not crawl whole sites, extract page text for LLMs, or host the assets it finds. Use for turning a single page URL into component code, brand identity, design tokens, and asset lists only.
The MiroMiro Design Extraction API turns a live website URL into design data you can build with. Its code operation resolves the page's real CSS cascade and returns a section such as the hero, pricing, or footer as a self-contained Tailwind, HTML and CSS, React, or Vue component with the tokens it uses. Other operations return the brand identity (color roles, palette, fonts, logos, and social profiles with a confidence score), the full design-token system as JSON, CSS custom properties, a Tailwind theme, or DTCG, and every image, SVG, font, video, audio file, and Lottie animation the page declares. Results are cached for 24 hours, and a render option runs the page's JavaScript in a real browser for client-rendered sites.
This API is usable in Jentic One now. Its AI-readiness score against Jentic's framework shows where it stands today and where improvements would make it even easier for agents to use.
Base layer of spec validity and structural soundness.
Aggregated quality score from linter diagnostics, weighted by severity.
Percentage of `$ref` references that resolve successfully.
Checks whether the API description parses successfully and conforms to its declared specification (e.g., OpenAPI).
Structural correctness score based on schema issues using logarithmic dampening.
Clarity, completeness, and ingestion readiness for developers and tooling.
How richly the API is illustrated with examples.
Percentage of examples that conform to their schemas.
Percentage of operations with complete response definitions (success, client error, server error).
Health of API ingestion, bundling, and resolution within Jentic pipelines.
Semantic breadth, depth, and agent comprehension for AI systems.
Coverage of descriptions across API elements.
Coverage of RFC 9457 Problem Details for error responses.
Coverage, uniqueness, and casing consistency of operationIds for AI inference.
Coverage of summaries across operations/tags/info.
Functional utility, complexity comfort, and AI orchestration readiness.
Agent comfort level based on API operational and structural complexity.
Trust, risk posture, and security compliance.
Average quality of security schemes based on authentication method strength (weakest link for OAuth2).
Findability, semantic richness, and reasoning readiness.
Clarity and depth of descriptions across API elements.
Score it yourself
Every API in the directory is allowlisted, so you can re-score it with no key required.
npx @jentic/api-scorecard-cli score <openapi-url>