Vap E

io.github.elestirelbilinc-sketch由 GitHub 帳號 elestirelbilinc-sketch 發布v1.14.1更新於 Oct 24, 2026

AI image, video & music generation. Flux, Veo 3.1, Suno V5. Free tier included.

已驗證Streamable HTTP可網頁執行媒體與設計AI 與機器學習

概覽

AI 產生的概覽

VAP 媒體 API 的遠端 MCP 伺服器，讓助理生成圖片、影片和音樂，並執行付費的影像與影片編輯。

可以做什麼
依文字提示生成圖片、影片和音樂，每次回傳任務 ID 供非同步追蹤，並可在執行前估算費用、查詢帳戶餘額。追蹤與列出生成任務和操作，完成後取得結果連結。以自然語言提示編輯圖片，支援局部重繪、去背和放大。裁剪並合併影片片段。
適用情境
適合希望由助理透過託管服務生成或編輯媒體、而不在本機執行模型，且能接受按次計費的情境。適用於快速創意草稿、圖片清理和簡單影片拼接。
執行需求
遠端端點，無需本機安裝套件。需要 VAP Media API 金鑰，透過 Authorization Bearer 標頭傳送（使用隨附本機代理時可用 VAP_API_KEY）。需要有余額的 VAP 帳戶，部分編輯工具要求 Tier 1 以上。
安裝前請注意
每次生成和編輯都會產生費用（例如每張圖片 0.18 美元、每支影片 1.96 美元、每首音樂 0.68 美元），請先查看餘額和估價。API 金鑰屬於憑證，勿放入共用設定。提示詞和來源媒體會傳送給第三方服務。生成任務為非同步執行，即使不取回結果也會扣費。

安裝

在 SourceWeft 中

  1. 開啟 儀表板中的 Vap E，將其新增到工作區。
  2. 為需要使用其工具的對話啟用該服務。

Web executable，透過 Streamable HTTP。 遠端服務在工作區中設定後即可從網頁執行環境執行。

交給代理程式安裝

把這段提示詞貼上到 Claude Code、Codex、Cursor 等代理程式，或 SourceWeft 對話中。代理程式會讀取該伺服器的安裝指南，先向你展示發布者、端點和登入方式，經你同意後再把它新增到你的客戶端。金鑰不會寫進任何命令。

閱讀 https://sourceweft.com/mcp/io.github.elestirelbilinc-sketch%2Fvap-e/install.md，並按其中說明把這個 MCP 伺服器接入我的客戶端。修改任何設定之前，先告訴我它的發布者、端點和登入方式，等我確認。不要把 API 金鑰寫進命令或我會提交的檔案；問我金鑰應從哪裡取得。

檢視代理程式所依據的安裝指南

其他 MCP 客戶端

把它新增到你客戶端的 mcpServers 設定中。

{
  "mcpServers": {
    "vap-e": {
      "type": "http",
      "url": "https://api.vapagent.com/mcp"
    }
  }
}

README

[VAP AI Saraswati emblem]

VAP AI

Agent-native AI platform for AI Room, Media API, and Coding Plan API.

Website | Developer Hub | AI Room | Dashboard | Status

Products

AI Room

Conversational creative workspace where custom-trained AI agents use session context to create images, video, voice, and music.

Media API

Unified generation API for image, video, and music workflows. Current public model surfaces include Pimo AI-Video, Aura Image Turbo, and Pira V5.5.

  • Developer Hub: https://vapagent.com/developer/
  • API base URL: https://api.vapagent.com/api/v1
  • Create generation: POST /api/v1/generations
  • Create operation: POST /api/v1/operations
  • Authentication: product-scoped VAP Media API key
  • MCP endpoint: https://api.vapagent.com/mcp

Coding Plan API

OpenAI-compatible API for coding agents, IDEs, editors, and automation workflows. Powered by Nemesis Deep Coder with model ID vap-code.

Start From The Product Surface

Use these entry points for new integrations:

NeedLink
Use AI Roomhttps://vapagent.com/new
Generate a Media API keyhttps://vapagent.com/developer/?key=media#keys
Generate a Coding Plan API keyhttps://vapagent.com/developer/?key=code#keys
View Coding Plan API planshttps://vapagent.com/new-dashboard/?billing=code#plans
Read Developer Hubhttps://vapagent.com/developer/
See Nemesis Deep Coderhttps://vapagent.com/models/nemesis-deep-coder.html

Media API Via MCP

This repository remains the public GitHub and MCP discovery surface for VAP Media API integrations. MCP is still supported for Claude Desktop, Claude Code, Cursor-compatible MCP clients, and other agent workflows that call VAP media tools.

Claude Desktop example:

json
{  "mcpServers": {    "vap": {      "url": "https://api.vapagent.com/mcp",      "headers": {        "Authorization": "Bearer YOUR_MEDIA_API_KEY"      }    }  }}

For clients that do not support headers directly, use the local proxy in mcp/vap_mcp_proxy.py and set VAP_API_KEY.

API Examples

Media API

bash
curl -X POST https://api.vapagent.com/api/v1/generations \  -H "Authorization: Bearer YOUR_MEDIA_API_KEY" \  -H "Content-Type: application/json" \  -d '{"kind":"image","prompt":"a neon city at night"}'

Coding Plan API

bash
curl -X POST https://api.vapagent.com/v1/chat/completions \  -H "Authorization: Bearer YOUR_CODING_PLAN_API_KEY" \  -H "Content-Type: application/json" \  -d '{"model":"vap-code","messages":[{"role":"user","content":"Review this patch."}]}'

Current Product Contracts

Use these current public contracts from Developer Hub:

  • Media API: https://api.vapagent.com/api/v1
  • Media generation: POST /api/v1/generations
  • Media operations: POST /api/v1/operations
  • MCP: https://api.vapagent.com/mcp
  • Coding Plan API: https://api.vapagent.com/v1

Discovery Files

  • server.json: MCP Registry server metadata
  • mcp.json: MCP package metadata
  • glama.json: Glama MCP listing metadata
  • mcp/tools.json: MCP tool schema metadata
  • mcp/vap_mcp_proxy.py: local stdio/http proxy for MCP clients

Links

License

MIT License. See LICENSE.

來源：README.md，提交 4274e2e

工具

16

generate_image

Generate an AI image from text prompt using VAP (Flux2 Pro). Returns a task ID for async tracking. Cost: $0.18

輸入參數prompt必填qualityaspect_ratio

get_task

Get the status and result of a generation task. Returns image URL when completed.

輸入參數task_id必填

list_tasks

List recent generation tasks with optional status filter.

輸入參數limitstatus

check_balance

Check VAP account balance. Returns available, reserved, and usable balances.

無需輸入參數。

estimate_cost

Estimate the cost of an image generation before executing. Cost: $0.18

輸入參數qualitynum_outputs

generate_video

Generate an AI video from text prompt using VAP (P-Video). Returns a task ID for async tracking. Cost: $1.96. IMPORTANT: Send ONLY the video description. Do NOT include any instructions, guidelines, or meta-text. Just the pure visual description.

輸入參數prompt必填durationresolutionaspect_ratiogenerate_audionegative_prompt

estimate_video_cost

Estimate the cost of a video generation. Cost: $1.96 (P-Video)

輸入參數durationresolutiongenerate_audio

generate_music

Generate AI music from text description using VAP. Returns a task ID for async tracking. Cost: $0.68. IMPORTANT: Send ONLY the music description. Do NOT include any instructions or meta-text. Describe: genre, mood, instruments, tempo, vocal style (or specify instrumental). Example prompt: "Upbeat indie folk song with acoustic guitar, warm vocals, and light percussion. Feel-good summer vibes.

輸入參數prompt必填durationaudio_formatinstrumentalloudness_preset

estimate_music_cost

Estimate the cost of music generation. Cost: $0.68

輸入參數durationaudio_format

inpaint

Remove or replace objects in an image using AI inpainting. Cost: $0.15. Requires Tier 1+.

輸入參數prompt必填mask_urlmedia_url必填

ai_edit

Edit images using AI with natural language prompts. Supports single and multi-image editing. Cost: $0.15. Requires Tier 1+.

輸入參數seedprompt必填media_url必填aspect_ratioadditional_images

background_remove

Remove the background from an image, leaving only the subject with transparency. Cost: $0.10. Requires Tier 1+.

輸入參數media_url必填

upscale

Upscale/enhance an image to higher resolution using AI. Cost: $0.15. Requires Tier 1+.

輸入參數scalemedia_url必填

video_trim

Trim a video to a specific time range. Cost: $0.05. Requires Tier 1+.

輸入參數end_time必填media_url必填start_time必填

video_merge

Merge multiple video clips into one continuous video. Cost: $0.05. Requires Tier 1+.

輸入參數media_urls必填

get_operation

Get the status and result of an operation. Returns output URL when completed.

輸入參數operation_id必填

版本歷史

1
  1. v1.14.1最新Sep 16, 2026

評分

已滿足適用檢查項的 100%

該評分衡量此收錄條目的完整程度，依據註冊中心條目、端點與公開來源計算得出。它不是對伺服器本身的評價，也不表示伺服器安全或目前可用：可連通性是帶日期的檢查結果，評分也不包含安裝量或使用者評價。

  • 近 30 天內可連通必需25 分
  • 已宣告認證方式必需10 分
  • 已列出工具必需15 分
  • 描述必需15 分
  • 圖示10 分
  • 文件15 分
  • 授權條款10 分