MCP MeasureMCP Measure

索引已观测

LMGram

Consent-based human matchmaking for peer advice, collaboration, and practical help.

我该使用它吗

质量与安全性

B
描述质量
93%
模式完整度
75%
命名质量
87%
投毒风险
80%
权限匹配度
100%
协议合规性
100%

发现（5）

  • HIGHTool poisoning patterns detected
  • MEDIUMTool 'list_messages' description contains placeholder text在 list_messages 中
  • LOWTool 'panel_snapshot' description lacks action verb在 panel_snapshot 中
  • LOWTool 'panel_register_device' description lacks action verb在 panel_register_device 中
  • INFOTool description contains placeholder or incomplete text在 list_messages 中

基于对工具定义和协议合规性的自动分析。

上下文开销

~2,707token 数（工具定义）
~1.1 KB典型响应大小
对注意力有显著影响（占 128k 上下文窗口的 2.11%）

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": {
    "lmgram": {
      "url": "https://api.lmgram.com/mcp"
    }
  }
}

远程端点

https://api.lmgram.com/mcpstreamable-http

它能做什么

工具清单

工具（17）

🟢 只读🟡 写入🔴 删除⚪ 未知
🟢get_current_user

Returns the LMGram profile linked to the connected account.

输入模式

{
  "type": "object",
  "properties": {},
  "additionalProperties": false
}
🟢request_human_match(topic, summary, need, matchType, urgency, ...)

Creates an opt-in LMGram human match request only after the user explicitly asks to use LMGram to find a collaborator, peer, mentor, or helper. Never invoke for general advice, dating or romantic matchmaking. Never include secrets, credentials, API keys, raw transcripts, customer data, or private documents; ask for a minimized non-sensitive summary first.

输入模式

{
  "type": "object",
  "properties": {
    "topic": {
      "type": "string",
      "description": "Short, non-sensitive title for the practical or professional session topic. Dating and romantic requests are not supported."
    },
    "summary": {
      "type": "string",
      "description": "Minimized non-sensitive summary. Never send secrets, credentials, API keys, raw transcripts, customer data, or private documents."
    },
    "need": {
      "type": "string",
      "description": "The concrete collaborator, peer, mentor, or helper the user explicitly wants. Dating and romantic matching are not supported."
    },
    "matchType": {
      "type": "string",
      "enum": [
        "peer",
        "mentor",
        "collaborator",
        "helper"
      ],
      "default": "peer"
    },
    "urgency": {
      "type": "number",
      "minimum": 0,
      "maximum": 1,
      "default": 0.5
    },
    "sensitivity": {
      "type": "string",
      "enum": [
        "normal",
        "sensitive",
        "high"
      ],
      "default": "normal"
    },
    "language": {
      "type": "string"
    },
    "maxCandidates": {
      "type": "integer",
      "minimum": 1,
      "maximum": 20,
      "default": 5
    }
  },
  "required": [
    "topic",
    "summary",
    "need"
  ],
  "additionalProperties": false
}
🟢list_match_requests(role, status, limit)

Lists recent match requests for the current LMGram user, optionally filtered by role and status.

输入模式

{
  "type": "object",
  "properties": {
    "role": {
      "type": "string",
      "enum": [
        "candidate",
        "requester"
      ],
      "default": "candidate"
    },
    "status": {
      "type": "string",
      "enum": [
        "pending",
        "chat_requested",
        "accepted",
        "declined"
      ],
      "description": "Optional exact request status to return."
    },
    "limit": {
      "type": "integer",
      "minimum": 1,
      "maximum": 50,
      "default": 10
    }
  },
  "additionalProperties": false
}
⚪start_chat(matchRequestId, body)

Starts a LMGram chat for a match request, or records that this side wants to chat.

输入模式

{
  "type": "object",
  "properties": {
    "matchRequestId": {
      "type": "string"
    },
    "body": {
      "type": "string",
      "description": "Optional first message."
    }
  },
  "required": [
    "matchRequestId"
  ],
  "additionalProperties": false
}
🟢list_messages(chatId)

Lists messages in a LMGram chat. End-to-end encrypted payloads are never exposed as ciphertext; they are represented by a safe placeholder.

输入模式

{
  "type": "object",
  "properties": {
    "chatId": {
      "type": "string"
    }
  },
  "required": [
    "chatId"
  ],
  "additionalProperties": false
}
🟡send_message(chatId, body)

Sends a message into a LMGram chat.

输入模式

{
  "type": "object",
  "properties": {
    "chatId": {
      "type": "string"
    },
    "body": {
      "type": "string"
    }
  },
  "required": [
    "chatId",
    "body"
  ],
  "additionalProperties": false
}
🟢open_lmgram_panel

Open LMGram chats, groups and suggestions for the current ChatGPT conversation. Each conversation has a separate context; never copy a transcript or private documents. This does not publish an intent or send a message.

输入模式

{
  "type": "object",
  "properties": {},
  "required": [],
  "additionalProperties": false
}

输出模式

{
  "type": "object",
  "properties": {
    "contextId": {
      "type": "string",
      "minLength": 43,
      "maxLength": 43
    }
  },
  "required": [
    "contextId"
  ],
  "additionalProperties": false
}
🟢panel_snapshot(contextId)

Refresh this conversation's private panel data.

输入模式

{
  "type": "object",
  "properties": {
    "contextId": {
      "type": "string",
      "minLength": 43,
      "maxLength": 43
    }
  },
  "required": [
    "contextId"
  ],
  "additionalProperties": false
}

输出模式

{
  "type": "object",
  "properties": {
    "contextId": {
      "type": "string",
      "minLength": 43,
      "maxLength": 43
    }
  },
  "required": [
    "contextId"
  ],
  "additionalProperties": false
}
🟢panel_chat(contextId, chatId, before)

Retrieve authorized chat data for client-side decryption, not model context.

输入模式

{
  "type": "object",
  "properties": {
    "contextId": {
      "type": "string",
      "minLength": 43,
      "maxLength": 43
    },
    "chatId": {
      "type": "string",
      "maxLength": 32
    },
    "before": {
      "type": "string",
      "maxLength": 32
    }
  },
  "required": [
    "contextId",
    "chatId"
  ],
  "additionalProperties": false
}

输出模式

{
  "type": "object",
  "properties": {
    "contextId": {
      "type": "string",
      "minLength": 43,
      "maxLength": 43
    }
  },
  "required": [
    "contextId"
  ],
  "additionalProperties": false
}
⚪panel_register_device(contextId, deviceId, publicKey)

Register only a public encryption key for this user's panel.

输入模式

{
  "type": "object",
  "properties": {
    "contextId": {
      "type": "string",
      "minLength": 43,
      "maxLength": 43
    },
    "deviceId": {
      "type": "string",
      "maxLength": 32
    },
    "publicKey": {
      "type": "string",
      "maxLength": 64
    }
  },
  "required": [
    "contextId",
    "deviceId",
    "publicKey"
  ],
  "additionalProperties": false
}

输出模式

{
  "type": "object",
  "properties": {
    "contextId": {
      "type": "string",
      "minLength": 43,
      "maxLength": 43
    }
  },
  "required": [
    "contextId"
  ],
  "additionalProperties": false
}
🟢panel_mark_read(contextId, chatId)

Mark the displayed chat as read.

输入模式

{
  "type": "object",
  "properties": {
    "contextId": {
      "type": "string",
      "minLength": 43,
      "maxLength": 43
    },
    "chatId": {
      "type": "string",
      "maxLength": 32
    }
  },
  "required": [
    "contextId",
    "chatId"
  ],
  "additionalProperties": false
}

输出模式

{
  "type": "object",
  "properties": {
    "contextId": {
      "type": "string",
      "minLength": 43,
      "maxLength": 43
    }
  },
  "required": [
    "contextId"
  ],
  "additionalProperties": false
}
🟡panel_sync_device_keys(contextId, chatId, sourceDeviceId, updates)

Share client-encrypted message keys only with this account's registered devices. Does not send messages, plaintext or private keys.

输入模式

{
  "type": "object",
  "properties": {
    "contextId": {
      "type": "string",
      "minLength": 43,
      "maxLength": 43
    },
    "chatId": {
      "type": "string",
      "maxLength": 32
    },
    "sourceDeviceId": {
      "type": "string",
      "maxLength": 32
    },
    "updates": {
      "type": "array",
      "minItems": 1,
      "maxItems": 60,
      "items": {
        "type": "object",
        "additionalProperties": false,
        "properties": {
          "messageId": {
            "type": "string",
            "maxLength": 32
          },
          "encryptionNonce": {
            "type": "string",
            "maxLength": 40
          },
          "encryptedKeys": {
            "type": "object",
            "maxProperties": 100,
            "additionalProperties": {
              "type": "string",
              "maxLength": 96
            }
          }
        },
        "required": [
          "messageId",
          "encryptionNonce",
          "encryptedKeys"
        ]
      }
    }
  },
  "required": [
    "contextId",
    "chatId",
    "sourceDeviceId",
    "updates"
  ],
  "additionalProperties": false
}

输出模式

{
  "type": "object",
  "properties": {
    "contextId": {
      "type": "string",
      "minLength": 43,
      "maxLength": 43
    }
  },
  "required": [
    "contextId"
  ],
  "additionalProperties": false
}
🟡panel_send_message(contextId, chatId, clientMessageId, body, encryptionScheme, ...)

Send a client-encrypted text message after the user presses Send. Retry only the identical payload and clientMessageId.

输入模式

{
  "type": "object",
  "properties": {
    "contextId": {
      "type": "string",
      "minLength": 43,
      "maxLength": 43
    },
    "chatId": {
      "type": "string",
      "maxLength": 32
    },
    "clientMessageId": {
      "type": "string",
      "maxLength": 36
    },
    "body": {
      "type": "string",
      "maxLength": 32000
    },
    "encryptionScheme": {
      "type": "string",
      "maxLength": 32
    },
    "encryptionNonce": {
      "type": "string",
      "maxLength": 40
    },
    "senderDeviceId": {
      "type": "string",
      "maxLength": 32
    },
    "encryptedKeys": {
      "type": "object",
      "additionalProperties": {
        "type": "string",
        "maxLength": 96
      },
      "maxProperties": 1000
    }
  },
  "required": [
    "contextId",
    "chatId",
    "clientMessageId",
    "body",
    "encryptionScheme",
    "encryptionNonce",
    "senderDeviceId",
    "encryptedKeys"
  ],
  "additionalProperties": false
}

输出模式

{
  "type": "object",
  "properties": {
    "contextId": {
      "type": "string",
      "minLength": 43,
      "maxLength": 43
    }
  },
  "required": [
    "contextId"
  ],
  "additionalProperties": false
}
⚪panel_start_chat(contextId, matchRequestId)

Request a chat with a suggestion in this context, without sending a plaintext introduction.

输入模式

{
  "type": "object",
  "properties": {
    "contextId": {
      "type": "string",
      "minLength": 43,
      "maxLength": 43
    },
    "matchRequestId": {
      "type": "string",
      "maxLength": 32
    }
  },
  "required": [
    "contextId",
    "matchRequestId"
  ],
  "additionalProperties": false
}

输出模式

{
  "type": "object",
  "properties": {
    "contextId": {
      "type": "string",
      "minLength": 43,
      "maxLength": 43
    }
  },
  "required": [
    "contextId"
  ],
  "additionalProperties": false
}
🔴panel_dismiss_suggestion(contextId, matchRequestId)

Decline an unopened match suggestion.

输入模式

{
  "type": "object",
  "properties": {
    "contextId": {
      "type": "string",
      "minLength": 43,
      "maxLength": 43
    },
    "matchRequestId": {
      "type": "string",
      "maxLength": 32
    }
  },
  "required": [
    "contextId",
    "matchRequestId"
  ],
  "additionalProperties": false
}

输出模式

{
  "type": "object",
  "properties": {
    "contextId": {
      "type": "string",
      "minLength": 43,
      "maxLength": 43
    }
  },
  "required": [
    "contextId"
  ],
  "additionalProperties": false
}
⚪panel_join_group(contextId, roomId)

Accept an existing group invitation related to this context.

输入模式

{
  "type": "object",
  "properties": {
    "contextId": {
      "type": "string",
      "minLength": 43,
      "maxLength": 43
    },
    "roomId": {
      "type": "string",
      "maxLength": 32
    }
  },
  "required": [
    "contextId",
    "roomId"
  ],
  "additionalProperties": false
}

输出模式

{
  "type": "object",
  "properties": {
    "contextId": {
      "type": "string",
      "minLength": 43,
      "maxLength": 43
    }
  },
  "required": [
    "contextId"
  ],
  "additionalProperties": false
}
🟡panel_create_intent(contextId, topic, summary, need, language)

Publish only the minimized intent explicitly reviewed and submitted by the user in this conversation panel. Do not include secrets, private transcripts, or dating requests.

输入模式

{
  "type": "object",
  "properties": {
    "contextId": {
      "type": "string",
      "minLength": 43,
      "maxLength": 43
    },
    "topic": {
      "type": "string",
      "maxLength": 240
    },
    "summary": {
      "type": "string",
      "maxLength": 2000
    },
    "need": {
      "type": "string",
      "maxLength": 1000
    },
    "language": {
      "type": "string",
      "maxLength": 40
    }
  },
  "required": [
    "contextId",
    "topic",
    "summary",
    "need"
  ],
  "additionalProperties": false
}

输出模式

{
  "type": "object",
  "properties": {
    "contextId": {
      "type": "string",
      "minLength": 43,
      "maxLength": 43
    }
  },
  "required": [
    "contextId"
  ],
  "additionalProperties": false
}

社区

评价此服务器

证据

最近观测

已验证未记录版本17 个工具
已验证未记录版本6 个工具
已验证未记录版本6 个工具