faktry

io.github.faktryv0.1.1更新于 Oct 6, 2026

Official faktry MCP server: generate and edit images, video, audio, and PDFs.

已验证Streamable HTTP可网页运行AI & MLFiles & StorageMedia & Design

概览

AI 生成的概览

为助手提供 70 多个 faktry 工具，用于生成和编辑图像、视频、音频、PDF 与 3D 素材，支持异步任务轮询和媒体库保存。

功能
把 faktry 媒体工具集接入 MCP 客户端：文生图、图像编辑、局部重绘、扩图、放大、去背景与合成；视频生成、数字人、运动控制、字幕与重新构图；语音合成、声音克隆、音乐与音效生成、转写与混音；PDF 创建与处理；以及从图像或文本生成 3D 素材。耗时任务会立即返回 job_id，需用 faktry_job_status 轮询直到得到 result_url。辅助工具可查询额度、浏览媒体库并保存结果以便后续会话复用。
适用场景
适合希望在聊天客户端内完成媒体制作的场景：产品图、带配音和字幕的 UGC 风格视频、声音克隆、PDF 拼装或 3D 素材。也适合已经使用 faktry 账户、并能接受异步任务轮询的工作流。
运行要求
可使用远程端点 faktry.ai/api/mcp，或通过 npx 运行 npm 桥接包 @faktry/mcp，后者需要 Node 18 及以上。认证方式为浏览器 OAuth，或通过 FAKTRY_API_KEY 环境变量、Authorization Bearer 请求头提供 API 密钥。需要 faktry 账户及额度，并需能访问该托管服务。
安装前请注意
使用会按 faktry 账户额度计费，生成和编辑调用都会消耗余额，可用 faktry_credits 查看。FAKTRY_API_KEY 密钥仅在创建时显示一次，应避免写入共享配置。声音克隆和数字人工具会处理声音或肖像数据，生成或编辑的媒体会上传到远程服务保存，包括存入媒体库的内容。

安装

在 SourceWeft 中

  1. 打开 控制台中的 faktry，将其添加到工作区。
  2. 为需要使用其工具的对话启用该服务。

Web executable，通过 Streamable HTTP。 远程服务在工作区中配置后即可从网页运行时运行。

其他 MCP 客户端

把它添加到你客户端的 mcpServers 配置中。

{
  "mcpServers": {
    "faktry-mcp": {
      "type": "http",
      "url": "https://faktry.ai/api/mcp"
    }
  }
}

README

[faktry]

faktry MCP

Bring the whole faktry media stack into the tools you already use.
Generate and edit images, video, audio, and PDFs from any MCP-compatible client.

[License: MIT] [Docs]

Install · Tools · Examples · Troubleshooting

[faktry platform]

The faktry MCP server exposes the full faktry toolkit — over 70 tools spanning image, video, audio, PDF, and 3D generation and editing — to any client that speaks the Model Context Protocol. Generate a product shot, cut it into a UGC-style video with a voiceover, subtitle it, and export a thumbnail sheet, all without leaving the chat.

  • [图片] Hosted, remote. Connect to https://faktry.ai/api/mcp — no local server to run. stdio-only clients can use the npx -y @faktry/mcp bridge.
  • OAuth or API key, your choice. One-click connect from Claude.ai, Claude Desktop, Cursor, and Hermes, or drop a static API key into any config-based client.
  • Async by default. Long-running jobs (video, some image/audio) return a job_id immediately; poll faktry_job_status until it's ready.
  • A library that remembers. Save any result with faktry_library_save and pull it back into a later session with faktry_library_list.
  • You pay through your faktry account. Usage draws down your existing credit balance — check it anytime with faktry_credits.

Install

One-click / OAuth

These clients handle authentication for you — paste the URL, sign in, approve.

MCP Server URL:

https://faktry.ai/api/mcp
Claude.ai
  1. Open Claude.ai → Settings → Integrations.
  2. Click Add MCP Server and paste the URL above.
  3. Claude.ai redirects you to faktry — sign in and click Approve.
  4. Done — Claude.ai stores the token automatically.
Claude Desktop
  1. Open Claude Desktop → Settings → Developer → MCP Servers.
  2. Click Add Server and paste the URL above.
  3. A browser window opens — sign in to faktry and click Approve.
  4. Restart Claude Desktop if tools don't appear immediately.
Cursor
  1. Open Cursor → Settings → AI → MCP Servers.
  2. Click Add MCP Server and paste the URL above.
  3. Cursor detects OAuth and opens a browser login window.
  4. Sign in to faktry, approve access, and return to Cursor.
Hermes Agent
  1. Open Hermes Agent → Settings → MCP Servers.
  2. Click + and paste the URL above.
  3. Hermes opens a browser login — sign in to faktry and approve.
  4. Toggle the server on — tools appear immediately.
ChatGPT Desktop (Business / Enterprise / Edu only)
  1. Enable Developer Mode in Settings → Beta Features.
  2. Go to Settings → MCP Servers → Add Server.
  3. Paste the URL above and follow the browser OAuth flow.

Config file / API key

Create an API key at faktry.ai/api/keys first, then add faktry to your client's config. Replace YOUR_API_KEY_HERE with your real key — it's only shown once when created.

Claude Code

Add to .mcp.json (project) or ~/.claude.json (user):

json
{  "mcpServers": {    "faktry": {      "type": "http",      "url": "https://faktry.ai/api/mcp",      "headers": { "Authorization": "Bearer YOUR_API_KEY_HERE" }    }  }}

Picked up automatically — no restart needed.

VS Code (Copilot)

Add to .vscode/mcp.json (workspace) or via Settings → MCP Servers:

json
{  "servers": {    "faktry": {      "type": "http",      "url": "https://faktry.ai/api/mcp",      "headers": { "Authorization": "Bearer YOUR_API_KEY_HERE" }    }  }}
Windsurf

Edit ~/.codeium/windsurf/mcp_config.json (%USERPROFILE%\.codeium\windsurf\mcp_config.json on Windows) — note Windsurf uses serverUrl, not url:

json
{  "mcpServers": {    "faktry": {      "serverUrl": "https://faktry.ai/api/mcp",      "headers": { "Authorization": "Bearer YOUR_API_KEY_HERE" }    }  }}

Restart Windsurf to activate the tools.

Continue.dev

Add to ~/.continue/config.yaml:

yaml
mcpServers:  faktry:    type: http    url: https://faktry.ai/api/mcp    requestOptions:      headers:        Authorization: "Bearer YOUR_API_KEY_HERE"

Continue.dev reloads automatically.

Codex CLI

Add to ~/.codex/config.toml:

toml
[mcp.servers.faktry]type = "http"url = "https://faktry.ai/api/mcp"
[mcp.servers.faktry.headers]Authorization = "Bearer YOUR_API_KEY_HERE"

Verify with codex mcp list.

stdio clients (npx) — Zed, older Claude Desktop configs, and anything without remote MCP

The @faktry/mcp package is a thin stdio bridge to the hosted server. Requires Node 18+.

Generic form:

bash
FAKTRY_API_KEY=YOUR_API_KEY_HERE npx -y @faktry/mcp

Claude Desktop (claude_desktop_config.json) and most other mcpServers clients:

json
{  "mcpServers": {    "faktry": {      "command": "npx",      "args": ["-y", "@faktry/mcp"],      "env": { "FAKTRY_API_KEY": "YOUR_API_KEY_HERE" }    }  }}

Zed (settings.json, %APPDATA%\Zed\settings.json on Windows):

json
{  "context_servers": {    "faktry": {      "source": "custom",      "command": "npx",      "args": ["-y", "@faktry/mcp"],      "env": { "FAKTRY_API_KEY": "YOUR_API_KEY_HERE" }    }  }}

Leave FAKTRY_API_KEY out to sign in through the browser (OAuth) instead.

Kilo Code
  1. Open the Kilo Code extension settings → MCP Servers.
  2. Add a new server using the same JSON snippet as Claude Code above.
  3. Replace YOUR_API_KEY_HERE with your actual key.
Other MCP clients

Add a remote HTTP server pointing at https://faktry.ai/api/mcp with header:

Authorization: Bearer YOUR_API_KEY_HERE

Full per-client snippets and paths are also generated at faktry.ai/mcp.

Tools

Your client picks which tool to call based on your prompt — you don't need to invoke them by name. faktry exposes 70+ tools across five categories:

CategoryExamplesWhat it covers
faktry_image_*generate, edit, inpaint, outpaint, upscale, removebackground, effects, collage, watermark, smartcropText-to-image, editing, upscaling, background removal, compositing
faktry_video_*text2video, image2video, avatar, motioncontrol, subtitle, reframe, stabilize, upscale, scenedetectGeneration, editing, avatars, motion control, subtitles, reframing
faktry_audio_*generatespeech, clonevoice, generatemusic, generatesound, mix, transcribe, trimSpeech synthesis, voice cloning, music/SFX generation, mixing
faktry_pdf_*create, merge, split, compress, extract, mdconvertPDF creation and manipulation
faktry_3d_*image23d, text23d3D asset generation from images or text

Plus utility tools:

ToolWhat it does
faktry_job_statusPoll an async job by job_id until it's completed, returning result_url
faktry_creditsCheck credit balance and current-month usage
faktry_library_listBrowse previously saved media assets
faktry_library_saveSave a result URL to the library for reuse across sessions

The async job pattern

Most generation tools (video, some image/audio) are async:

  1. Call the generation tool → receive a job_id.
  2. Poll faktry_job_status with that job_id every few seconds.
  3. When status is completed, the response contains result_url.
  4. Optionally persist it with faktry_library_save.

Example prompts

Try these once faktry is connected:

Generate a studio product shot of a matte black wireless speaker on awhite background, then upscale it to 4K.
Turn this product image into a 9:16 UGC-style video with a voiceoverreading the description, and add subtitles.
Remove the background from this photo, then composite it onto agradient backdrop.
Clone this voice sample and generate a 30-second narration readingthis script.
Check my faktry credits, then show me what's in my library.

Troubleshooting

Tools don't appear after connecting
  • Claude.ai / Desktop: Settings → Integrations (or Developer → MCP Servers), confirm faktry shows as connected. Remove and re-add if it failed silently — make sure pop-ups aren't blocked so the OAuth window can open.
  • Claude Code: claude mcp list should show the server.
  • Codex: codex mcp list should show faktry. Start a new session after adding.
  • Quick smoke test: ask your client "check my faktry credits".
Authentication or API key errors
  • For config-file clients, confirm YOUR_API_KEY_HERE was replaced with a real, active key from faktry.ai/api/keys — keys are shown only once at creation.
  • For OAuth clients, disconnect and reconnect the connector to redo the flow.
A generation keeps loading

Video renders and some audio/image jobs take longer than a plain image generation. The tool returns a pending job_id immediately — keep polling faktry_job_status until it reports completed.

Links

License

MIT © faktry

来源：README.md，提交 6411bc2

工具

0
工具元数据尚未被收录。

版本历史

1
  1. v0.1.1最新Oct 6, 2026