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)
Image infrastructure for agents. Upload, transform, and deliver images through MCP, CLI, or API.
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.
AuthorizationValues marked secret are keys to your account. Never share them or paste them into a chat.
You need
Which app do you use?
Pick your app above. We remember it, so next time the button says “Install in …”.
This address needs a key in a request header, which the add-connector menu cannot set. Use Claude Code, Cursor or VS Code instead.
Run this in your terminal
claude mcp add --transport http pixelvault https://mcp.pixelvault.dev/mcp --header 'Authorization: Bearer {pixelvault_api_key}'Fill in the empty values before running it.
Paste into .cursor/mcp.json in your project, or ~/.cursor/mcp.json
{
"mcpServers": {
"pixelvault": {
"url": "https://mcp.pixelvault.dev/mcp",
"headers": {
"Authorization": "Bearer {pixelvault_api_key}"
}
}
}
}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": {
"pixelvault": {
"type": "http",
"url": "https://mcp.pixelvault.dev/mcp",
"headers": {
"Authorization": "Bearer {pixelvault_api_key}"
}
}
}
}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.
This address needs a key in a request header, which the add-connector menu cannot set. Use Claude Code, Cursor or VS Code instead.
Your choice stays on this device in a cookie; no account needed.
Authorization secret required HTTP headerThese settings are generated from the maker’s registry entry. If the README says otherwise, follow the README.
Tool names and descriptions exactly as the server reported them on Nov 7, 2026. The maker wrote them, not us; we show them as plain text.
get_imageGet metadata (CDN URL, size, MIME type, dimensions) for one PixelVault image by id. Maps to GET /v1/images/:id. Requires a PixelVault API key sent as a Bearer token in the Authorization header. The returned CDN URL supports on-the-fly transforms via query params — e.g. ?w=400&fit
list_imagesList images in your PixelVault project, most recent first, optionally filtered by custom metadata or AI tag, or searched by AI description (q: alt text, caption, OCR text, best match first). Maps to GET /v1/images with pagination. Requires a PixelVault API key sent as a Bearer to
get_usageShow your plan, this month's usage (storage, uploads) and what is left of the quota, so you can check before a bulk upload. Maps to GET /v1/usage. Read-only. Requires a PixelVault API key sent as a Bearer token in the Authorization header.
delete_imagePermanently delete one PixelVault image by id. Maps to DELETE /v1/images/:id. Requires a PixelVault API key sent as a Bearer token in the Authorization header.
update_image_metadataReplace one image's custom metadata (a JSON object you can later filter list_images by). The object replaces the stored one, so send every key to keep; null clears it. Maps to PATCH /v1/images/:id. Requires a PixelVault API key sent as a Bearer token in the Authorization header.
list_image_tagsList the AI tags on your described images with how many images carry each, most used first (at most 40). Pass one to list_images (tag) to see those images. Maps to GET /v1/images/tags. Read-only. Requires a PixelVault API key sent as a Bearer token in the Authorization header.
describe_imageStart the AI description (alt text, caption, OCR text, up to 6 tags) of an image you already uploaded without describe=true. Idempotent: an image that is already described or in progress returns its state and uses none of your allowance. It takes a few seconds: call get_image unt
create_upload_urlPrepare a public image upload from a local shell file when no host attachment download URL exists. Compute the file size and SHA-256 in the shell, then call this tool with filename, size and sha256. Returns a 10-minute URL and curl command to PUT the raw file bytes without an API
upload_imageUpload an image publicly to PixelVault and get an instant CDN URL. Maps to POST /v1/images. Provide exactly one of `file` (a host-provided attachment), `source_url` (a public http(s) URL the server fetches), or `data` (base64-encoded bytes). Prefer file for attachments; optional
upload_batchUpload many images (1–50) in one call, grouped into a collection — e.g. a CI run or a set of generated variants. Maps to POST /v1/images/batch. Each item is `data` (base64) or `source_url` (server-fetched, SSRF-guarded), with optional `filename`/`metadata`. Set `visibility: "priv
list_collectionsList your project's collections (CI runs, generation sets, albums — what upload_batch creates), newest first. Maps to GET /v1/collections. Optional `type` and `name` filters (name is the CI run id, generation id, etc.) and pagination. Requires a PixelVault API key sent as a Beare
get_collectionGet one collection and its images by id. Maps to GET /v1/collections/:id. Each image has a `visibility`: for `private` images the returned URL is the plain CDN URL, which returns 403 without a signature, so call sign_url with the image id for a viewable link (a signed URL cannot
close_collectionClose a collection (e.g. when a PR merges). Maps to POST /v1/collections/:id/close. Call get_collection first and check its `type`: for `ci_build` collections closing makes every image in it eligible for deletion, which a background sweep (hourly, in batches) then carries out per
sign_urlMint a time-limited signed URL for a private image. Maps to POST /v1/images/:id/sign-url. Provide `id` and optional `expires_in` (seconds, default 3600). The signature binds the image, so the URL can't be replayed against another image; strip it and the CDN returns 403. Deleting
transform_imageBuild an on-the-fly transform URL for a PixelVault image (resize, crop, format/quality, AI background removal, blur/sharpen/rotate/flip, brightness/contrast/saturation, and same-project watermark tiling). Provide exactly one of `url` (a PixelVault CDN URL) or `id` (an image id, r
rescue_imgurScan a web page for hotlinked Imgur images and return a PixelVault rescue URL for each — a proxy that lazily rehosts the image on first request, keeping it working through the UK Imgur block. The anonymous rescue tier is a durable ~30-day edge cache (see `cache_ttl_days` in the r
generate_imageGenerate an image from a text prompt with a Cloudflare Workers AI model and host it on PixelVault in one call; returns the permanent CDN URL. Maps to POST /v1/images/generate. Spends credits from the monthly grant (free 10, Starter 300, Pro 1500; resets on the 1st, UTC): flux-1-s
upgrade_planStart a paid-plan checkout (`starter` or `pro`). Maps to POST /v1/billing/checkout. Returns a Stripe Checkout URL — it does NOT charge anything: the account owner must open the URL and pay, so show it only to them. The plan changes only after Stripe confirms, so call get_usage af
manage_billingOpen the Stripe billing portal to change plan, update payment details or cancel. Maps to POST /v1/billing/portal. Returns a URL for the account owner to open; nothing changes until they act there. The URL can cancel the subscription and change payment details, so show it only to
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 publisher (pixelvault-dev) is not the repo owner (pixelvault-dev); someone else may have listed it
We have not read this repo from GitHub yet
We have not read this repo from GitHub yet
Responded in 142 ms (HTTP 200)
No sign-in needed
Read 19 tool descriptions and found no hidden instructions to the AI
Compared with every active server in this category in the registry. Stars count only when the publisher owns the repo.
It passed 2 of 3 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.
You need to fill in: Authorization
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 (0.4.0) reached the registry 14 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.
No README found in the repo