lens

Create your digital twin: a public page that answers from your own material, with citations.

我该使用它吗

质量与安全性

A
描述质量
100%
模式完整度
93%
命名质量
100%
投毒风险
100%
权限匹配度
100%
协议合规性
100%

基于对工具定义和协议合规性的自动分析。

上下文开销

~1,027token 数(工具定义)
~3.4 KB典型响应大小
对注意力有中等影响(占 128k 上下文窗口的 0.80%)

这是每次将服务器的工具加载到模型上下文窗口时所消耗的大致 token 数。数值越高,可用于其他任务的注意力就越少。

安装

一键安装

将以下内容添加到你的 `claude_desktop_config.json` 文件中:

{
  "mcpServers": {
    "lens": {
      "url": "https://app.mindola.ai/mcp/lens"
    }
  }
}

远程端点

https://app.mindola.ai/mcp/lensstreamable-http

它能做什么

工具清单

工具(3)

🟢 只读🟡 写入🔴 删除⚪ 未知
🟡create_digital_twin_lens(displayName, tagline, bio, context, audience, ...)

Create a shareable Mindola lens (AI clone) from a profile the user has approved in this conversation. Call this only after summarizing the proposed lens for the user and receiving explicit confirmation. Use only information the user provided or confirmed in the conversation; do not claim access to hidden memory, private chats, or external accounts.

输入模式

{
  "type": "object",
  "properties": {
    "displayName": {
      "type": "string",
      "description": "The user's preferred public name for the lens, after they confirm it.",
      "maxLength": 80
    },
    "tagline": {
      "type": "string",
      "description": "A short one-line summary of what this lens represents.",
      "maxLength": 160
    },
    "bio": {
      "type": "string",
      "description": "A concise first-person or third-person biography assembled from user-confirmed information.",
      "maxLength": 2000
    },
    "context": {
      "type": "string",
      "description": "Extra grounding context for how the lens should answer. Use only information the user confirmed.",
      "maxLength": 4000
    },
    "audience": {
      "type": "string",
      "description": "Who this lens is intended to help or answer questions for.",
      "maxLength": 500
    },
    "sourceSummary": {
      "type": "string",
      "description": "Short note describing what ChatGPT used to assemble the profile, without exposing hidden chain-of-thought or unapproved private data.",
      "maxLength": 1500
    },
    "topicName": {
      "type": "string",
      "description": "The Mindola space name that will hold the seeded facts.",
      "maxLength": 80
    },
    "locale": {
      "type": "string",
      "description": "Optional BCP-47 locale for user-facing lens copy.",
      "maxLength": 32
    },
    "tone": {
      "type": "string",
      "enum": [
        "warm",
        "neutral",
        "professional",
        "playful"
      ],
      "description": "The visitor-facing tone of the lens."
    },
    "accentColor": {
      "type": "string",
      "description": "Optional hex color, for example #4f46e5.",
      "pattern": "^#[0-9a-fA-F]{6}$"
    },
    "expertise": {
      "type": "array",
      "description": "Specific domains, skills, or interests the user confirmed.",
      "maxItems": 12,
      "items": {
        "type": "string",
        "maxLength": 160
      }
    },
    "boundaries": {
      "type": "array",
      "description": "Things the lens should not claim, discuss, or answer beyond. Ask the user to confirm these before calling.",
      "maxItems": 12,
      "items": {
        "type": "string",
        "maxLength": 160
      }
    },
    "facts": {
      "type": "array",
      "description": "Concrete user-approved facts to seed the lens memory. Each item should stand alone.",
      "maxItems": 30,
      "items": {
        "type": "string",
        "maxLength": 320
      }
    },
    "suggestedPrompts": {
      "type": "array",
      "description": "Starter questions visitors can ask the lens.",
      "maxItems": 4,
      "items": {
        "type": "string",
        "maxLength": 120
      }
    },
    "vibeWords": {
      "type": "array",
      "description": "Up to four short adjectives for the lens profile.",
      "maxItems": 4,
      "items": {
        "type": "string",
        "maxLength": 24
      }
    }
  },
  "additionalProperties": false
}

输出模式

{
  "type": "object",
  "properties": {
    "displayName": {
      "type": "string"
    },
    "topicName": {
      "type": "string"
    },
    "shareUrl": {
      "type": "string"
    },
    "claimUrl": {
      "type": [
        "string",
        "null"
      ]
    },
    "manageUrl": {
      "type": [
        "string",
        "null"
      ]
    },
    "claimRequired": {
      "type": "boolean"
    },
    "expiresAt": {
      "type": [
        "string",
        "null"
      ]
    },
    "factCount": {
      "type": "number"
    },
    "suggestedPrompts": {
      "type": "array",
      "items": {
        "type": "string"
      }
    }
  },
  "required": [
    "displayName",
    "topicName",
    "shareUrl",
    "claimUrl",
    "manageUrl",
    "claimRequired",
    "expiresAt",
    "factCount",
    "suggestedPrompts"
  ],
  "additionalProperties": false
}
🟡add_lens_knowledge(lens_token, items)

Add text notes or URLs to a Mindola lens created in this conversation (before it is claimed). Content runs through the standard ingest pipeline so the lens can answer from it, with citations.

输入模式

{
  "type": "object",
  "properties": {
    "lens_token": {
      "type": "string",
      "description": "The token from the lens's share URL (the part after /me/)."
    },
    "items": {
      "type": "array",
      "maxItems": 20,
      "items": {
        "type": "object",
        "properties": {
          "type": {
            "type": "string",
            "enum": [
              "text",
              "url"
            ]
          },
          "content": {
            "type": "string",
            "maxLength": 20000
          },
          "title": {
            "type": "string",
            "maxLength": 80
          }
        },
        "required": [
          "type",
          "content"
        ],
        "additionalProperties": false
      }
    }
  },
  "required": [
    "lens_token",
    "items"
  ],
  "additionalProperties": false
}
🟢get_lens_status(lens_token)

Check a Mindola lens: its share URL, whether it has been claimed, and how much of its knowledge has finished ingesting (ready / pending / failed).

输入模式

{
  "type": "object",
  "properties": {
    "lens_token": {
      "type": "string",
      "description": "The token from the lens's share URL (the part after /me/)."
    }
  },
  "required": [
    "lens_token"
  ],
  "additionalProperties": false
}

推荐提示词

retrieve_data
Get details about [item] from lens
预期工具: get_lens_status
fetch_info
Fetch [information type] using lens
预期工具: get_lens_status

社区

评价此服务器

证据

最近观测

已验证未记录版本3 个工具
已验证未记录版本3 个工具