Pixly

它能帮你做什么

我们还没有为它写通俗的介绍。下面是我们从注册中心读取到的事实。

• 连接方式: 在云端运行
• 传输方式: Streamable HTTP
• 分类: 设计、图片与音频

安装

选择你使用的 AI 应用，然后复制设置。

你需要

• 在云端运行无需安装任何东西，通过互联网连接

你用的是哪个应用？

https://pixly.app/api/mcp

claude mcp add --transport http pixly https://pixly.app/api/mcp

{
  "mcpServers": {
    "pixly": {
      "url": "https://pixly.app/api/mcp"
    }
  }
}

{
  "servers": {
    "pixly": {
      "type": "http",
      "url": "https://pixly.app/api/mcp"
    }
  }
}

https://pixly.app/api/mcp

你的选择通过 Cookie 保存在这台设备上，无需账户。

这些设置是根据作者在注册中心的条目生成的。如果 README 的写法不同，请以 README 为准。

它能做什么

工具名称和描述与服务器在 2026年11月8日 报告的完全一致。这些内容由作者撰写，不是我们写的；我们以纯文本形式显示。

• virtual_staging

  AI virtual staging: furnish and style an empty (or badly furnished) room photo in a chosen interior design style — also the tool for "redesign this room" or "show this room in another style". HD, watermark-free results. Costs credits from the user's Pixly balance.

• edit_staged_photo

  Change one thing in a photo and keep everything else exactly as it is — the parts the edit doesn't touch are copied from the photo, not redrawn. Works on any photo: a staged room, any Pixly result, or the user's own picture (swap the sofa, recolour a wall, remove a car, change th

• cinematic_motion

  Photo to video: turn a single listing photo into a short cinematic real estate video with a professional camera move (zoom, orbit, fly-through, crane up…). No second frame needed. Costs credits from the user's Pixly balance.

• photo_to_video

  Turn a start photo — and optionally an end photo — into a video. With an end photo the camera travels from the first photo and lands exactly on the second (e.g. two views of the same kitchen → an orbit around the island). Pick a ready-made camera move or write your own prompt; ch

• before_after_reel

  Before and after video: animate a before→after transformation (e.g. empty room → staged, or a renovation, or a daylight photo → its Day to Night version with videoIntent day_to_night) into a social-ready reveal reel from two photos. Both photos are required. Only have one? Make t

• flythrough_video

  Flythrough video: one continuous camera move through a home from 2-4 room photos in walking order. Each pair of neighbouring photos becomes a 3-second move that flies from one room into the next, and the moves are stitched into one film with a cinematic speed ramp. Returns one jo

• before_after_image

  Free (no credits): one comparison image from a before and an after photo of the same place — e.g. an empty room and its virtual staging, a daylight shot and its dusk version, a room before and after a restyle. Side by side, stacked, or split by a line; original shape, square, 4:5

• watermark_photos

  Free (no credits): put the agent's or agency's mark on listing photos — text (name, phone) and/or the logo saved on their Pixly account — in a corner or the middle, up to 10 photos per call. Returns every marked photo at once.

• blur_faces_plates

  Free (no credits): find and blur people's faces and car licence plates in property photos before they're posted — street and exterior shots, open houses. Automatic detection, 1–5 photos per call. Says so when nothing is found.

• straighten_photo

  Free (no credits): automatically straighten a property photo — leaning walls made vertical (converging verticals from a tilted camera) and a tilted horizon levelled, like a pro photographer's lens correction. Crops the edges slightly. Best before staging or any other edit. Says s

• draw_floor_plan

  Free (no credits): draw a clean 2D floor plan from a layout YOU write — read the user's description, or look at their hand sketch or brochure plan, and send every room as a rectangle in metres (or feet) plus doors and windows. Returns the drawn plan (areas and dimensions on it) a

• enhance_photo

  Real estate photo editing in one step: turn an amateur listing photo into a finished, professional photo (exposure, color, clarity). Returns 3 variants to pick from. Costs credits from the user's Pixly balance.

• declutter_photo

  Remove clutter, mess, and personal items from a listing photo while keeping the room, furniture, and architecture intact. Costs credits from the user's Pixly balance.

• remove_furniture

  Remove furniture from a photo: empty a room and rebuild the surfaces behind what was taken out, keeping the architecture and camera angle identical — the blank canvas for restaging. By default it also removes the room's own joinery (fitted kitchen, built-in wardrobes, bathroom fi

• day_to_night

  Day to dusk (virtual twilight): turn a daytime listing photo into a magazine-style evening scene — lights on, warm glow, dusk sky — with the building and camera angle unchanged. Costs credits from the user's Pixly balance.

• replace_sky

  Replace a grey, overcast or blown-out sky in an exterior listing photo with a natural clear blue sky, matching the light and keeping the building, trees and reflections unchanged. A run takes about 1–2 minutes; poll get_job until it completes. Costs credits from the user's Pixly

• replace_lawn

  Turn patchy, brown or bare grass in an exterior listing photo into a healthy green lawn, leaving driveways, paths, beds and the house untouched. Costs credits from the user's Pixly balance.

• restyle_room

  Give one or more surfaces of a room a new finish — repaint the walls, change the floor, reface the kitchen cabinets and worktop, retile the bathroom or refinish the furniture — while the room's layout, fittings, windows and camera angle stay exactly as shot. Costs credits from th

• reset_season

  Photograph the same property in another season: take the snow off the ground, the roof and the branches and put the garden in leaf, or move a summer shot to autumn or winter. The house, the driveway, the paths and the camera angle stay exactly as shot. Costs credits from the user

• makeover_exterior

  Design the garden of an exterior listing photo as one coherent scheme — planting, a terrace or deck, a pool, a pergola, a fire pit, a fence, lighting — in a chosen style, at the right scale and perspective, while the house, its roof, the driveway, the sun direction and the neighb

• render_floor_plan

  Turn a picture of any floor plan — a hand sketch, a photo of a brochure page, a blueprint, a screenshot — into a polished real-estate floor plan: a furnished top-down 2D plan with real floors and room names, or a 3D isometric cutaway. The rooms, walls, doors and windows are kept

• touch_up_exterior

  One-click retouch of an exterior listing photo: a natural sky where it is dull, a healthy lawn where the grass is dead, a clean driveway, and bins, hoses and other clutter removed — each only where needed, with the house unchanged. Costs credits from the user's Pixly balance.

• upscale_hd

  Enlarge a photo by up to four times with real detail, to a maximum of 4096 px on the long edge — for print, MLS uploads and large screens. Nothing in the photo changes. Free on any result Pixly made (pass its r2Path from list_library); 1 credit on your own photo. Costs credits fr

• plot_sign

  Place a photorealistic 3D monument sign with your exact text (price, area, SOLD) onto a photo of an empty plot, matched to perspective and lighting. Costs credits from the user's Pixly balance.

• upload_image_from_url

  Rarely needed: generation tools take an imageUrl directly. Use this only to import a photo into Pixly ahead of time, e.g. to reuse one URL across several tools without refetching it each time.

• create_upload_ticket

  Get a presigned upload URL for a LOCAL image file — the one case a URL can't cover. PUT the file to uploadUrl with the same Content-Type, then pass the returned r2Path to a generation tool instead of imageUrl. Ticket expires in 1 hour.

• get_job

  Check a generation job's status and fetch result URLs when it's done. Video jobs take 1-5 minutes — poll this with a few seconds between calls. For several jobs at once (a plan's steps), pass jobIds.

• list_library

  List the user's recent generations, or with type "uploads" their own uploaded source photos, newest first. Uploads are the way in when the photo is a local file: connectors cannot send bytes, so the user adds it at https://pixly.app/app and you pick up the r2Path here.

• view_pixly_photo

  Look at one of the user's Pixly photos or videos (a video gives its first frame) — returns a small copy of the picture you can see. Use it before writing anything about a Pixly result: a caption, a listing description, what changed. You cannot see Pixly results otherwise. Free; c

• get_credit_balance

  The user's current Pixly credit balance and plan.

• list_listings

  The user's property folders (a listing such as "12 Oak St" holds that property's photos, staged versions, edits, videos and floor plan): id, title, how many photos and videos. Use when the user names a property, or before work for one, so results land in its folder.

• create_listing

  Create a property folder before work on a whole property (a set of listing photos, a walkthrough). Ask the user once which property it is and pass that as title; if they skip, leave title out. Pass the returned listingId to the tools so their results land in it. Costs nothing.

• update_listing

  Rename one of the user's property folders, only when they ask.

• plan_photos

  Several listing photos at once: one plan of photo steps (straighten, enhance, sky, lawn, staging, declutter…) with one total and one confirmation, instead of a confirm per photo. The first call runs nothing: it returns each step's cost, the total and a confirmationCode, and the P

• get_checkout_link

  Use when a tool fails for lack of credits or a plan. Returns a Stripe Checkout URL for the user to pay on — this never charges anything by itself, and you cannot pay for them. With no active plan it is a subscription (plans: starter (Starter: $5.99/week, 20 credits); agent (Agent

基础检查i

我们检查的是任何人都能看到的信息；我们没有阅读代码。不适用的检查项不计入总数。 · 检查于 2026年11月9日

• ! 注意

  作者与仓库一致

  注册中心没有列出源代码仓库，因此无法将发布者（pixly.app）与代码对应起来

• … 尚未检查

  仍在维护

  我们还没有从 GitHub 读取这个仓库

• … 尚未检查

  有许可证

  我们还没有从 GitHub 读取这个仓库

• ✓ 通过

  端点有响应

  在 821 毫秒内响应（HTTP 200）

• – 不适用

  使用 OAuth 登录

  无需登录

• ✓ 通过

  工具描述中没有隐藏指令

  已读取 35 个工具描述，未发现针对 AI 的隐藏指令

与“设计、图片与音频”中的其他服务器比较i

• 星标数多于此分类中其他 2802 个服务器的 0%
• 发布到注册中心的时间比此分类中其他 2802 个服务器的 41% 更新
• 基础检查结果等于或优于此分类中其他 2802 个服务器的 44%

与注册中心里这个分类中所有活跃的服务器比较。只有当发布者拥有该仓库时，星标数才计入。

替代选择

同一分类中其他作者的服务器，按星标数排序。

如何看卡片：↓ 每周下载量i★ GitHub 星标i三大支柱：评价 · 检查 · 使用情况i更新 = 最新版本发布到注册中心的时间i

作者的 README

仓库中没有找到 README