feat(vscode): bring face + fish into hosted-HTTP codegen

face/README.md

@@ -47,18 +47,18 @@ mcp-face-transform
 mcp-face-transform --transport http --port 8000
 ```
 
-## Available Tools
-
-| Tool | Endpoint | Purpose |
-|------|----------|---------|
-| `face_detect_keypoints` | `POST /face/analyze` | Detect 90+ keypoints per face |
-| `face_beautify` | `POST /face/beautify` | Smoothing / whitening / slimming / eye enlarging |
-| `face_change_age` | `POST /face/change-age` | Age or de-age a portrait |
-| `face_change_gender` | `POST /face/change-gender` | Swap perceived facial gender characteristics |
-| `face_swap` | `POST /face/swap` | Move source face onto target image |
-| `face_cartoonize` | `POST /face/cartoon` | Convert portrait to cartoon style |
-| `face_detect_liveness` | `POST /face/detect-live` | Detect live vs printed/screen face |
-| `face_get_usage_guide` | _client-side_ | Concise tool usage reference |
+## Tool Reference
+
+| Tool | Description |
+|------|-------------|
+| `face_detect_keypoints` | Detect 90+ keypoints per face (multi-face supported). |
+| `face_beautify` | Smoothing, whitening, face slimming, and eye enlarging. |
+| `face_change_age` | Age or de-age a portrait. |
+| `face_change_gender` | Swap perceived facial gender characteristics. |
+| `face_swap` | Move a source face onto a target image (with optional async webhook). |
+| `face_cartoonize` | Render a portrait in cartoon / animated style. |
+| `face_detect_liveness` | Distinguish a live capture from a printed / screen photo. |
+| `face_get_usage_guide` | Concise client-side tool usage reference. |
 
 ## Example
 

face/vscode/.vscodeignore

@@ -1,3 +1,11 @@
-.github/**
+.vscode/**
+.vscode-test/**
+node_modules/**
+*.vsix
 .git/**
 .gitignore
+.eslintrc.json
+.prettierrc
+**/*.map
+**/tsconfig.json
+**/.editorconfig

face/vscode/README.md

@@ -1,49 +1,119 @@
 # Face Transform MCP
 
-Face analysis & transformation — keypoints, beautify, age/gender, swap, cartoon, liveness — via the Ace Data Cloud Face API.
+Face analysis and transforms — keypoints, beautify, age/gender, swap, cartoon, liveness. (Alpha)
 
-This extension provides a [Model Context Protocol (MCP)](https://modelcontextprotocol.io/) server for VS Code, enabling AI assistants like GitHub Copilot to call the Face Transform API directly.
+[![VS Code Marketplace](https://img.shields.io/visual-studio-marketplace/v/acedatacloud.mcp-face-transform?label=VS%20Code)](https://marketplace.visualstudio.com/items?itemName=acedatacloud.mcp-face-transform) [![PyPI](https://img.shields.io/pypi/v/mcp-face-transform.svg?label=PyPI)](https://pypi.org/project/mcp-face-transform/) [![Hosted MCP](https://img.shields.io/badge/hosted-mcp-blue)](https://face.mcp.acedata.cloud/mcp)
 
-## Features
+Bring AceDataCloud's Face Transform APIs into Copilot Chat. Detect 90+ keypoints per face, beautify portraits, age or de-age, swap perceived gender, face-swap between photos, cartoonize, and detect liveness.
 
-- **8 tools** covering the full AceDataCloud Face API surface
-- Zero-install: Uses `uvx` for automatic package management
-- Works with GitHub Copilot, Claude, and other MCP-compatible AI assistants
+This extension registers the **face** MCP server with VS Code so GitHub
+Copilot and any other agent that speaks the [Model Context Protocol](https://modelcontextprotocol.io/)
+can call it directly from chat.
 
-## Prerequisites
+---
 
-1. **Python 3.10+** with `uvx` (from [uv](https://github.com/astral-sh/uv)) installed
-2. **Ace Data Cloud API Token** — Get one at [platform.acedata.cloud](https://platform.acedata.cloud)
+## Quick Start
 
-## Setup
+1. **Install this extension.** VS Code registers the `face` MCP server automatically.
+2. **Get an API token** from [Ace Data Cloud](https://platform.acedata.cloud) → *API Keys*. New accounts include free trial credit.
+3. **Open Copilot Chat** in agent mode and ask for a image task — VS Code will prompt for the token the first time and store it securely.
 
-1. Install this extension from the VS Code Marketplace
-2. When prompted, enter your Ace Data Cloud API token
-3. The MCP server will be available to AI assistants automatically
+> The default config talks to the **hosted streamable-HTTP endpoint** at
+> `https://face.mcp.acedata.cloud/mcp` — no Python, no `uvx`, no local install needed.
 
-You can also manually configure the token in your VS Code settings (`.vscode/mcp.json`):
+### Example prompts
 
-```json
+- "Detect every face in https://example.com/group.jpg and return their keypoints."
+- "Beautify https://example.com/me.jpg with smoothing 15 and whitening 25."
+- "Swap the face from https://example.com/headshot.jpg onto https://example.com/scene.jpg."
+
+---
+
+## Tool Reference
+
+**8 tools** available via this server.
+
+| Tool | Description |
+| --- | --- |
+| `face_detect_keypoints` | Detect 90+ keypoints per face (multi-face supported). |
+| `face_beautify` | Smoothing, whitening, face slimming, and eye enlarging. |
+| `face_change_age` | Age or de-age a portrait. |
+| `face_change_gender` | Swap perceived facial gender characteristics. |
+| `face_swap` | Move a source face onto a target image (with optional async webhook). |
+| `face_cartoonize` | Render a portrait in cartoon / animated style. |
+| `face_detect_liveness` | Distinguish a live capture from a printed / screen photo. |
+| `face_get_usage_guide` | Concise client-side tool usage reference. |
+
+## Pricing
+
+All Face APIs are currently in Alpha. Free trial credit on sign-up. See full pricing at [https://docs.acedata.cloud](https://docs.acedata.cloud).
+
+---
+
+## Configuration
+
+This extension contributes the following entry to your VS Code MCP config:
+
+```jsonc
 {
   "servers": {
     "face": {
+      "type": "http",
+      "url": "https://face.mcp.acedata.cloud/mcp",
+      "headers": { "Authorization": "Bearer ${input:acedatacloud_api_token}" }
+    }
+  },
+  "inputs": [
+    {
+      "type": "promptString",
+      "id": "acedatacloud_api_token",
+      "description": "Ace Data Cloud API token",
+      "password": true
+    }
+  ]
+}
+```
+
+VS Code will prompt for the token on first use and persist it in the OS
+secret store (Keychain / Credential Manager / libsecret).
+
+### Alternative: local stdio (no network roundtrip)
+
+If you prefer running the server locally — for offline dev, air-gapped
+environments, or to pin to a specific PyPI version — install
+[`uv`](https://docs.astral.sh/uv/) and replace your `mcp.json` entry with:
+
+```jsonc
+{
+  "servers": {
+    "face": {
+      "type": "stdio",
       "command": "uvx",
       "args": ["mcp-face-transform"],
-      "env": {
-        "ACEDATACLOUD_API_TOKEN": "your-token-here"
-      }
+      "env": { "ACEDATACLOUD_API_TOKEN": "${input:acedatacloud_api_token}" }
     }
   }
 }
 ```
 
+`uvx` will download and run the latest [`mcp-face-transform`](https://pypi.org/project/mcp-face-transform/) on demand.
+
+### Alternative: OAuth via Dynamic Client Registration
+
+The hosted endpoint also accepts OAuth 2.1 with [DCR](https://datatracker.ietf.org/doc/html/rfc7591).
+Drop the `headers` and `inputs` blocks and VS Code will run the auth flow on
+first use (redirect URL `http://127.0.0.1:33418` or `https://vscode.dev/redirect`).
+
+---
+
 ## Links
 
-- [PyPI Package](https://pypi.org/project/mcp-face-transform/)
-- [Source Code](https://github.com/AceDataCloud/FaceTransformMCP)
-- [Ace Data Cloud Platform](https://platform.acedata.cloud)
-- [API Documentation](https://docs.acedata.cloud)
+- **Hosted endpoint:** https://face.mcp.acedata.cloud/mcp
+- **PyPI package:** [`mcp-face-transform`](https://pypi.org/project/mcp-face-transform/)
+- **Source repository:** https://github.com/AceDataCloud/FaceTransformMCP
+- **Ace Data Cloud platform:** https://platform.acedata.cloud
+- **MCP documentation:** https://docs.acedata.cloud
 
 ## License
 
-MIT
+MIT — see [LICENSE](LICENSE).

face/vscode/package.json

@@ -1,8 +1,8 @@
 {
   "name": "mcp-face-transform",
   "displayName": "Face Transform MCP",
-  "description": "Face analysis & transformation (keypoints, beautify, age/gender, swap, cartoon, liveness) via Ace Data Cloud API",
-  "version": "0.1.0",
+  "description": "Face analysis and transforms — keypoints, beautify, age/gender, swap, cartoon, liveness. (Alpha)",
+  "version": "0.2.0",
   "publisher": "acedatacloud",
   "icon": "icon.png",
   "license": "MIT",
@@ -20,11 +20,21 @@
     "copilot",
     "ace data cloud",
     "face",
+    "face detection",
+    "keypoints",
     "face swap",
-    "face transform",
-    "liveness detection",
-    "beautify"
+    "beautify",
+    "cartoon",
+    "liveness",
+    "portrait",
+    "age",
+    "gender",
+    "transform"
   ],
+  "galleryBanner": {
+    "color": "#0E1117",
+    "theme": "dark"
+  },
   "homepage": "https://github.com/AceDataCloud/FaceTransformMCP",
   "repository": {
     "type": "git",
@@ -36,14 +46,20 @@
   "contributes": {
     "mcpServers": {
       "face": {
-        "command": "uvx",
-        "args": [
-          "mcp-face-transform"
-        ],
-        "env": {
-          "ACEDATACLOUD_API_TOKEN": ""
+        "type": "http",
+        "url": "https://face.mcp.acedata.cloud/mcp",
+        "headers": {
+          "Authorization": "Bearer ${input:acedatacloud_api_token}"
         }
       }
-    }
+    },
+    "inputs": [
+      {
+        "type": "promptString",
+        "id": "acedatacloud_api_token",
+        "description": "Ace Data Cloud API token. Sign in at https://platform.acedata.cloud and create a token in 'API Keys'. Stored securely by VS Code.",
+        "password": true
+      }
+    ]
   }
 }

fish/README.md

@@ -40,7 +40,7 @@ mcp-fish
 mcp-fish --transport http --port 8000
 ```
 
-## Available Tools
+## Tool Reference
 
 | Tool | Description |
 |------|-------------|

fish/vscode/.vscodeignore

@@ -1,3 +1,11 @@
-.github/**
+.vscode/**
+.vscode-test/**
+node_modules/**
+*.vsix
 .git/**
 .gitignore
+.eslintrc.json
+.prettierrc
+**/*.map
+**/tsconfig.json
+**/.editorconfig