Figma Context MCP
Give your coding agent access to your Figma data. Implement designs in any framework in one-shot.
Description from the maker (original English)
Mobile UI/UX design research: real app screens, user flows and UI patterns with cited evidence.
These three are never merged into one score: opinions, measurements and popularity are different things.
We have not written a plain-language summary for this one yet. Below are the facts we read from the registry.
Pick the AI app you use, then copy the settings.
You need
Which app do you use?
Pick your app above. We remember it, so next time the button says “Install in …”.
Works in Claude on the web, desktop and mobile:
https://agent.flownato.com/mcpRun this in your terminal
claude mcp add --transport http flownato https://agent.flownato.com/mcpPaste into .cursor/mcp.json in your project, or ~/.cursor/mcp.json
{
"mcpServers": {
"flownato": {
"url": "https://agent.flownato.com/mcp"
}
}
}Your browser asks to open the app, then the app shows the settings and asks you to confirm. Key values are left empty for you to fill in.
Paste into .vscode/mcp.json in your project
{
"servers": {
"flownato": {
"type": "http",
"url": "https://agent.flownato.com/mcp"
}
}
}Your browser asks to open the app, then the app shows the settings and asks you to confirm. Key values are left empty for you to fill in.
ChatGPT connects only to servers on the internet:
https://agent.flownato.com/mcpMenu names can differ slightly by plan and over time.
Your choice stays on this device in a cookie; no account needed.
These settings are generated from the maker’s registry entry. If the README says otherwise, follow the README.
We have not read this server’s tool list yet; we check endpoints in rotation.
We check information anyone can see; we have not read the code. Checks that do not apply are left out of the count. · Checked Nov 7, 2026
The registry name was verified against the domain flownato.com (the registry checks domain ownership)
We have not read this repo from GitHub yet
We have not read this repo from GitHub yet
Not checked yet; we check endpoints in rotation
Not checked yet; we check endpoints in rotation
Not checked yet; we read the tool list when we check the endpoint
Compared with every active server in this category in the registry. Stars count only when the publisher owns the repo.
It passed 1 of 1 applicable basic checks. We have not read the code. Before adding a key or token, see what access the server asks for and use a key with only the permissions you need.
The registry does not list a required key, though some servers ask you to sign in on first use.
It runs in the maker’s cloud. Connect over the internet with nothing to install, but what you send passes through the maker’s servers.
The GitHub repo last changed –, and the latest version (1.0.0) reached the registry 36 days ago.
Apps that support MCP, such as Claude Desktop, Claude Code, Cursor and VS Code. It connects by URL, so apps that accept remote MCP servers work too, such as ChatGPT in developer mode.
In the same category: Figma Context MCP, Claude Real Video, Figma MCP Server
Servers in the same category from other makers, sorted by stars.
Reading a card:★ GitHub starsiThree pillars: reviews · checks · usageiUpdated = latest version published to the registryi
Give your coding agent access to your Figma data. Implement designs in any framework in one-shot.
Let any LLM watch a video locally — and search everything it has ever watched.
Figma’s official server gives AI your design data so it can turn designs into code or describe them accurately. Connects to Figma’s cloud.
Copied from the maker’s repo as written. PopNx MCP has not reviewed or endorsed it. HTML, images and scripts are removed. Links lead off this site.
A remote, read-only Model Context Protocol server for mobile UI/UX design research. Built for product designers and developers, it gives MCP-compatible assistants (Claude, Cursor, Codex, Windsurf and others) access to Flownato's library of real mobile app screens, user flows and UI patterns, with every answer tied to published screen evidence.
https://agent.flownato.com/mcp (Streamable HTTP)Flownato records real mobile apps such as Swiggy, Zepto, BookMyShow, redBus and Nykaa, then publishes reviewed screens, complete journeys and a UI pattern taxonomy. The MCP server lets an assistant search and compare that evidence instead of guessing what an interface looks like.
Typical questions it can answer with citations:
| Tool | Purpose |
|---|---|
whoami | Reports the authenticated membership, what that tier can see, and the client. |
search_ui_library | Finds concrete journey examples for a research question. |
search_ui_screens | Screen-level search with exact interface text matching. |
analyze_ui_pattern | Measures one screen pattern, UI element, flow pattern or flow action across an optional app or category scope. |
explore_ui_patterns | Discovers recurring patterns when you do not yet know which one to analyze. |
compare_ui_flows | Compares the same product moment across two or three named apps. |
inspect_app_library | Summarizes one app's published coverage, patterns, elements and journeys. |
check_ui_presence | Scoped yes/no presence checks with evidence. |
inspect_library_coverage | Paginated inventory of what is published. |
inspect_ui_resources | Exact release-bound follow-up retrieval for returned resourceRef objects. |
get_screen_image | Returns a published screen image inline after another tool supplies its asset ID. |
All tools are read-only. Results cite the exact journey and step identifiers they were built from, and pattern counts are descriptive, never presented as best practice.
The plugin installs the MCP server together with the flownato-ui-research skill, which tells the assistant when to look up real app flows and how to cite them:
/plugin marketplace add flownato/flownato-mcp
/plugin install flownato@flownato
The same repository carries plugin manifests for Cursor (.cursor-plugin/) and Codex (.codex-plugin/). The skill lives in skills/flownato-ui-research/SKILL.md.
In clients that accept a remote mcpServers configuration, use:
{
"mcpServers": { "flownato": { "type": "http", "url": "https://agent.flownato.com/mcp" } } }
Claude Code:
claude mcp add --transport http flownato https://agent.flownato.com/mcp
On first use the client opens https://api.flownato.com/oauth/authorize in your browser. Sign in with your Flownato account and approve read-only access. Tokens are short lived and refresh silently until you revoke the grant.
Discovery documents:
https://agent.flownato.com/.well-known/oauth-protected-resource/mcphttps://api.flownato.com/.well-known/oauth-authorization-server| Plan | What the MCP returns |
|---|---|
| Not signed in | The endpoint answers 401 with OAuth discovery metadata, so your client opens Flownato sign-in. Creating an account is free. |
| Free account | All eleven tools. Journeys marked free on flownato.com (up to three per app) return every screen. Other journeys return their first 3 screens; the rest come back with locked: true, no asset ID and no image URL. Every result carries an access block with the locked counts and an upgrade link, and get_screen_image answers PREMIUM_REQUIRED for a locked screen. |
| Premium | All eleven tools with the complete library. |
Both tiers are metered at a default safety limit of 500 tool calls per UTC day.
Prompts, arguments and results are not stored. Screenshot URLs are signed and short lived, and the image endpoint rechecks membership when the image is fetched.
The server reads only the published product catalog: reviewed, immutable releases with privacy redaction applied. It has no access to raw crawl data, model output or other users' accounts. It answers UI and product-design research questions and declines unrelated requests.
This repository documents the hosted service and holds its plugin manifests and skill. The server itself is part of Flownato's private product codebase (Python, FastAPI) and is not open source.
lhm.plugin.json declares the existing LobeHub listing's category, remote endpoint, and all eleven MCP tool schemas. Its category is developer (displayed as Developer Skills on LobeHub), reflecting its use for UI/UX design and product development.
The endpoint requires OAuth before protocol initialization and tool discovery. An unauthenticated request returning 401 is expected; use the protected-resource metadata above to start sign-in. Directory crawlers cannot discover the tool list from an unauthenticated connection alone.
The server currently exposes tools, with no MCP prompt templates or MCP resources. Returned resourceRef objects are arguments to inspect_ui_resources; they are not entries in the MCP resources/list capability. Keep marketplace declarations consistent with the server's actual capabilities.
The public badge above supports LobeHub's README claim-status check. Prefer this public verification route when available. Updating this public listing does not justify granting a directory read/write access to private GitHub repositories.
Already verified maintainers can update the existing listing with the official @lobehub/market-cli (Node.js 22 or newer):
# Sign in to LobeHub, then check existing GitHub verification:
npx -y @lobehub/market-cli login npx -y @lobehub/market-cli github status
npx -y @lobehub/market-cli plugin update --dir "$PWD"
Review requested permissions before connecting any GitHub account. If a claim or update requires broad repo or workflow access, leave that connection disabled and request a narrower verification method from LobeHub.
When tool signatures change, refresh tools from the authenticated server's tools/list response before publishing. Keep prompts and resources empty unless the server implements them. The listing version remains 1.0.0 for this metadata correction; it does not represent a new server release. Verify the public listing after each update.