MeatSpace
Human-in-the-loop for AI agents. Submit choices, get a human decision.
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": {
"meatspace": {
"url": "https://meatspace.run/api/mcp"
}
}
}遠端端點
https://meatspace.run/api/mcpstreamable-http它能做什麼
工具清單
工具（3）
🟢get_service_status
Check whether MeatSpace is available and when to use a human. Call this when deciding whether to escalate a subjective or high-consequence choice.
輸入結構描述
{
"type": "object",
"properties": {},
"required": []
}⚪ask_human(agent_name, title, content, content_type, choices, ...)
Present content to a human and ask them to choose between options. Use this for subjective judgment, approval, preference, or tie-breaks. Avoid using it for deterministic checks or reversible low-stakes choices. The tool waits briefly for a result, then returns pending if the human has not responded yet.
輸入結構描述
{
"type": "object",
"properties": {
"agent_name": {
"type": "string",
"description": "Your agent/tool name (max 100 chars)"
},
"title": {
"type": "string",
"description": "Short title for the request (max 200 chars)"
},
"content": {
"type": "string",
"description": "Content for the human to review (text, markdown, HTML, or image URL). Max 50KB."
},
"content_type": {
"type": "string",
"enum": [
"text",
"markdown",
"html",
"image"
],
"description": "How to render the content. Default: text"
},
"choices": {
"type": "array",
"items": {
"type": "object",
"properties": {
"id": {
"type": "string",
"description": "Unique choice identifier (max 50 chars)"
},
"label": {
"type": "string",
"description": "Human-readable label (max 100 chars)"
}
},
"required": [
"id",
"label"
]
},
"minItems": 2,
"maxItems": 4,
"description": "2-4 choices for the human to pick from"
},
"callback_url": {
"type": "string",
"description": "Optional HTTPS webhook URL for async notification"
},
"metadata": {
"type": "object",
"description": "Optional metadata passed through to webhook"
},
"decision_reason": {
"type": "string",
"description": "Why the agent is escalating this to a human (max 500 chars)"
},
"confidence": {
"type": "number",
"description": "Agent confidence between 0 and 1"
},
"consequence_of_wrong_choice": {
"type": "string",
"description": "Why a wrong choice matters (max 500 chars)"
},
"recommended_option": {
"type": "string",
"description": "Optional choice id the agent currently recommends"
},
"run_id": {
"type": "string",
"description": "Optional workflow run identifier"
},
"trace_id": {
"type": "string",
"description": "Optional trace identifier"
},
"timeout_seconds": {
"type": "number",
"description": "Request expiry in seconds (default 3600, max 86400)"
}
},
"required": [
"agent_name",
"title",
"choices"
]
}🟡provision_api_key(name, email)
Create a MeatSpace API key instantly. No authentication required. Returns a Bearer token for use with the ask_human tool. Max 5 active keys per email address.
輸入結構描述
{
"type": "object",
"properties": {
"name": {
"type": "string",
"description": "Agent or tool name (max 100 chars)"
},
"email": {
"type": "string",
"description": "Owner email address"
}
},
"required": [
"name",
"email"
]
}社群
證據