Resemble AI
Resemble AI: deepfake detection, media intelligence, and watermarking actions, plus docs tools.
我該用這個嗎
品質與安全性
根據工具定義與協定合規性的自動化分析。
上下文成本
這是每次將伺服器的工具載入模型上下文時所消耗的約略 token 數量。數量越高,可用於其他工作的注意力就越少。
安裝
一鍵安裝
將以下內容加入你的 `claude_desktop_config.json` 檔案:
{
"mcpServers": {
"resemble-mcp": {
"url": "https://mcp.resemble.ai/mcp"
}
}
}遠端端點
https://mcp.resemble.ai/mcpstreamable-httphttps://mcp.resemble.ai/ssesse它能做什麼
工具清單
工具(6)
🟢resemble_docs_lookup(topic)
Get comprehensive documentation about a Resemble AI topic. This is the PRIMARY tool - use it first for any question about Resemble AI. Returns complete, aggregated documentation in a single call. Supports flexible matching - use topic IDs, names, or keywords: Topics (use these IDs or related keywords): - voice-cloning: clone, custom voice, train, recordings, dataset - text-to-speech: tts, synthesize, generate speech, audio generation - speech-to-speech: s2s, voice conversion - speech-to-text: stt, transcribe, transcription - getting-started: auth, api key, quickstart, rate limit - voice-design: design voice, generate voice, candidate - streaming: websocket, real-time, low latency - detect: deepfake, watermark, identity, source tracing, intelligence - agents: conversational, phone, call, webhook, knowledge base - sdks: python, node, npm, pip, library - projects-clips: project, clip, manage - audio-tools: enhance, edit, improve audio - ssml: markup, prosody, emphasis, phoneme - custom-pronunciations: pronunciation, phoneme, bulk, zip - duets: multi-speaker, conversation, dialogue, voice pair - account: billing, usage, team, subscription - voice-settings: preset, pace, temperature, pitch, exaggeration - term-substitutions: term, substitution, replacement Examples: - "voice-cloning" → Everything about cloning voices - "watermark" → Watermarking docs (matched via 'detect' topic) - "tts" → Text-to-speech documentation - "pronunciation" → Custom pronunciations documentation - "duets" → Multi-speaker conversational audio
輸入結構描述
{
"type": "object",
"properties": {
"topic": {
"type": "string",
"description": "Topic ID, name, or keyword (e.g., 'voice-cloning', 'watermark', 'tts', 'transcribe')"
}
},
"required": [
"topic"
]
}🟢resemble_search(query, limit)
Search across all Resemble AI documentation. Use this when you need to find specific information not covered by topic lookup, or when exploring what documentation exists. Returns matching pages with snippets.
輸入結構描述
{
"type": "object",
"properties": {
"query": {
"type": "string",
"description": "Search query (e.g., 'webhook callback', 'rate limits', 'emotion parameter')"
},
"limit": {
"type": "integer",
"description": "Maximum results to return",
"default": 5
}
},
"required": [
"query"
]
}🟢resemble_get_page(path)
Get the full content of a specific documentation page. Use this when you know the exact page path from a previous search.
輸入結構描述
{
"type": "object",
"properties": {
"path": {
"type": "string",
"description": "Page path (e.g., 'getting-started/quickstart', 'voice-creation/voices/create')"
}
},
"required": [
"path"
]
}🟢resemble_list_topics
List all available documentation topics with descriptions. Use this to discover what topics are available.
輸入結構描述
{
"type": "object",
"properties": {}
}🟢resemble_api_endpoint(method, path)
Get detailed OpenAPI specification for a specific API endpoint. Returns request/response schemas, parameters, and examples. Use this when you need exact API details for implementation. Examples: - method: "POST", path: "/synthesize" → TTS synthesis endpoint - method: "GET", path: "/voices" → List voices endpoint - method: "POST", path: "/speech-to-text" → Transcription endpoint
輸入結構描述
{
"type": "object",
"properties": {
"method": {
"type": "string",
"description": "HTTP method (GET, POST, PUT, PATCH, DELETE)",
"enum": [
"GET",
"POST",
"PUT",
"PATCH",
"DELETE"
]
},
"path": {
"type": "string",
"description": "API endpoint path (e.g., '/synthesize', '/voices', '/projects')"
}
},
"required": [
"method",
"path"
]
}🟢resemble_api_search(query)
Search the OpenAPI specification for endpoints matching a query. Use this to find which API endpoints are available for a feature. Examples: - "voice" → All voice-related endpoints - "streaming" → Streaming endpoints - "agent" → Agent-related endpoints
輸入結構描述
{
"type": "object",
"properties": {
"query": {
"type": "string",
"description": "Search query for API endpoints"
}
},
"required": [
"query"
]
}社群
證據