MCP MeasureMCP Measure

インデックス観測済み

MCP Verify

Trust, freshness, policy, and discovery layer for public MCP servers.

使うべきか

品質と安全性

A
説明の品質
96%
スキーマの完全性
77%
命名の品質
95%
ポイズニングのリスク
100%
権限の一致
100%
プロトコルへの準拠
100%

検出事項（1）

  • LOWTool 'recommend_servers' description lacks action verbrecommend_servers 内

ツール定義とプロトコルへの準拠に関する自動分析に基づいています。

コンテキストコスト

~2,537トークン数（ツール定義）
~894 B一般的なレスポンスサイズ
注意への影響は中程度（128k コンテキストの 1.98%）

これは、サーバーのツールがモデルのコンテキストに読み込まれるたびに消費されるおおよそのトークン数です。数が多いほど、ほかのタスクに使える注意が減ります。

インストール

ワンクリックインストール

これを `claude_desktop_config.json` ファイルに追加してください:

{
  "mcpServers": {
    "verify": {
      "url": "https://verify.sentinelsignal.io/mcp"
    }
  }
}

リモートエンドポイント

https://verify.sentinelsignal.io/mcpstreamable-http

できること

ツール一覧

ツール（22）

🟢 読み取り専用🟡 書き込み🔴 削除⚪ 不明
🟢search(query, limit)

ChatGPT-compatible read-only search alias. Returns MCP server results with id, title, and url.

入力スキーマ

{
  "type": "object",
  "properties": {
    "query": {
      "type": "string",
      "description": "Search query such as healthcare MCP, web search MCP, ChatGPT-compatible MCPs, or OAuth MCP servers."
    },
    "limit": {
      "type": "integer",
      "minimum": 1,
      "maximum": 20,
      "description": "Maximum number of search results to return."
    }
  },
  "required": [
    "query"
  ]
}
🟢fetch(id)

ChatGPT-compatible read-only fetch alias. Returns a full MCP server item with id, title, text, url, and metadata.

入力スキーマ

{
  "type": "object",
  "properties": {
    "id": {
      "type": "string",
      "description": "Canonical MCP server identifier in namespace/name format."
    }
  },
  "required": [
    "id"
  ]
}
🟢search_servers(capabilities, auth_preference, client_target, risk_tolerance, query, ...)

Search and rank MCP servers by capability, auth preference, client target, and risk tolerance.

入力スキーマ

{
  "type": "object",
  "properties": {
    "capabilities": {
      "type": "array",
      "items": {
        "type": "string"
      },
      "description": "Normalized capability taxonomy terms such as healthcare, search, read, exec, files, oauth, or prompts."
    },
    "auth_preference": {
      "type": "string",
      "enum": [
        "any",
        "oauth",
        "unauthenticated"
      ],
      "description": "Preferred auth mode. Use unauthenticated when the agent must avoid login flows."
    },
    "client_target": {
      "type": "string",
      "enum": [
        "openai_connectors",
        "claude_desktop",
        "smithery",
        "generic_http",
        "healthcare"
      ],
      "description": "Optional target client or integration surface."
    },
    "risk_tolerance": {
      "type": "string",
      "enum": [
        "any",
        "low",
        "medium",
        "high"
      ],
      "description": "How much tool-surface risk is acceptable."
    },
    "query": {
      "type": "string",
      "description": "Optional free-text query to bias ranking toward a specific use case or domain."
    },
    "limit": {
      "type": "integer",
      "minimum": 1,
      "maximum": 20,
      "description": "Maximum number of ranked candidates to return."
    }
  }
}
🟢recommend_servers(task, capabilities, auth_preference, client_target, risk_tolerance, ...)

Recommend MCP servers for a plain-language task and return ranked matches with install config.

入力スキーマ

{
  "type": "object",
  "properties": {
    "task": {
      "type": "string",
      "description": "Plain-language task description such as 'I need an MCP for healthcare denial scoring in OpenAI connectors'."
    },
    "capabilities": {
      "type": "array",
      "items": {
        "type": "string"
      },
      "description": "Optional explicit capability constraints that override or extend the inferred task capabilities."
    },
    "auth_preference": {
      "type": "string",
      "enum": [
        "any",
        "oauth",
        "unauthenticated"
      ],
      "description": "Preferred auth mode."
    },
    "client_target": {
      "type": "string",
      "enum": [
        "openai_connectors",
        "claude_desktop",
        "smithery",
        "generic_http",
        "healthcare"
      ],
      "description": "Optional target client or integration surface."
    },
    "risk_tolerance": {
      "type": "string",
      "enum": [
        "any",
        "low",
        "medium",
        "high"
      ],
      "description": "How much tool-surface risk is acceptable."
    },
    "limit": {
      "type": "integer",
      "minimum": 1,
      "maximum": 20,
      "description": "Maximum number of recommendations to return."
    }
  },
  "required": [
    "task"
  ]
}
🟢get_server_report(identifier)

Return the full machine-readable verify report for a specific MCP server identifier.

入力スキーマ

{
  "type": "object",
  "properties": {
    "identifier": {
      "type": "string",
      "description": "Canonical server identifier in namespace/name format."
    }
  },
  "required": [
    "identifier"
  ]
}
🟢compare_servers(identifiers)

Compare up to four MCP servers side by side across score, verdict, auth, tool count, prompts/resources, and freshness.

入力スキーマ

{
  "type": "object",
  "properties": {
    "identifiers": {
      "type": "array",
      "items": {
        "type": "string"
      },
      "minItems": 2,
      "maxItems": 4,
      "description": "Canonical server identifiers in namespace/name format."
    }
  },
  "required": [
    "identifiers"
  ]
}
🟢route_task(task, candidate, capabilities, client_target, max_risk, ...)

Choose the best MCP server/tools for a task. Use decide_agent_call when evaluating one specific attempted tool call.

入力スキーマ

{
  "type": "object",
  "properties": {
    "task": {
      "type": "string",
      "description": "Task the agent wants to perform."
    },
    "candidate": {
      "type": "string",
      "description": "Optional namespace/name server identifier to evaluate directly."
    },
    "capabilities": {
      "type": "array",
      "items": {
        "type": "string"
      },
      "description": "Optional capability constraints such as healthcare, search, database, read, or write."
    },
    "client_target": {
      "type": "string",
      "enum": [
        "openai_connectors",
        "claude_desktop",
        "smithery",
        "generic_http",
        "healthcare"
      ]
    },
    "max_risk": {
      "type": "string",
      "enum": [
        "low",
        "medium",
        "high",
        "critical"
      ],
      "description": "Maximum allowed tool risk."
    },
    "requires_oauth": {
      "type": "boolean"
    },
    "no_write_without_approval": {
      "type": "boolean"
    },
    "max_freshness_hours": {
      "type": "number"
    }
  },
  "required": [
    "task"
  ]
}
⚪decide_agent_call(agent_id, user_id, server_id, tool, action, ...)

Verify Gateway decision point. Decide whether this agent may call this tool, with this payload, for this user, right now. Writes an audit-ledger decision event.

入力スキーマ

{
  "type": "object",
  "properties": {
    "agent_id": {
      "type": "string"
    },
    "user_id": {
      "type": "string"
    },
    "server_id": {
      "type": "string",
      "description": "Server identifier in namespace/name format."
    },
    "tool": {
      "type": "string"
    },
    "action": {
      "type": "string",
      "enum": [
        "invoke",
        "read",
        "write",
        "preview",
        "list",
        "connect"
      ]
    },
    "mandate_id": {
      "type": "string",
      "description": "Optional Verify Mandate key or id."
    },
    "payload_classification": {
      "type": "array",
      "items": {
        "type": "string"
      }
    },
    "requested_cost": {
      "type": "number"
    },
    "max_cost": {
      "type": "number"
    },
    "approval_token_present": {
      "type": "boolean"
    },
    "client": {
      "type": "string"
    },
    "max_risk": {
      "type": "string",
      "enum": [
        "low",
        "medium",
        "high",
        "critical"
      ]
    },
    "requires_oauth": {
      "type": "boolean"
    },
    "no_write_without_approval": {
      "type": "boolean"
    },
    "max_freshness_hours": {
      "type": "number"
    },
    "max_runtime_seconds": {
      "type": "integer"
    },
    "audit_context": {
      "type": "object"
    }
  },
  "required": [
    "agent_id",
    "user_id",
    "server_id",
    "tool"
  ]
}
🟡create_mandate(user_id, agent_id, server_id, scope, budget_limit, ...)

Create a Verify Mandate: signed delegated authority for an agent/user/server scope, budget, validity window, and tool constraints.

入力スキーマ

{
  "type": "object",
  "properties": {
    "user_id": {
      "type": "string"
    },
    "agent_id": {
      "type": "string"
    },
    "server_id": {
      "type": "string"
    },
    "scope": {
      "type": "string"
    },
    "budget_limit": {
      "type": "number"
    },
    "valid_until": {
      "type": "string",
      "format": "date-time"
    },
    "requires_human_approval": {
      "type": "boolean"
    },
    "signature": {
      "type": "string"
    },
    "constraints": {
      "type": "object"
    },
    "metadata": {
      "type": "object"
    }
  },
  "required": [
    "user_id",
    "agent_id",
    "scope"
  ]
}
🟢get_evidence_pack(server_id, agent_id, user_id, limit)

Return Verify Ledger evidence packs with invocations, hash-chained audit events, decisions, outcomes, mandates, and cost summaries.

入力スキーマ

{
  "type": "object",
  "properties": {
    "server_id": {
      "type": "string"
    },
    "agent_id": {
      "type": "string"
    },
    "user_id": {
      "type": "string"
    },
    "limit": {
      "type": "integer",
      "minimum": 1,
      "maximum": 1000
    }
  }
}
⚪compile_docs(title, source_type, source_url, openapi_spec, markdown_docs, ...)

Generate agent-readable MCP artifacts from product docs: server card, llms.txt, tool schemas, examples, policy metadata, score preview, and hosted docs page.

入力スキーマ

{
  "type": "object",
  "properties": {
    "title": {
      "type": "string"
    },
    "source_type": {
      "type": "string",
      "enum": [
        "openapi",
        "github_repo",
        "markdown",
        "postman",
        "website",
        "pdf",
        "help_center",
        "mixed"
      ]
    },
    "source_url": {
      "type": "string"
    },
    "openapi_spec": {
      "type": "object"
    },
    "markdown_docs": {
      "type": "string"
    },
    "postman_collection": {
      "type": "object"
    },
    "website_docs": {
      "type": "string"
    },
    "server_identifier": {
      "type": "string"
    },
    "policy_metadata": {
      "type": "object"
    },
    "metadata": {
      "type": "object"
    }
  },
  "required": [
    "title"
  ]
}
🟢export_policy(identifier, client, max_risk, requires_oauth, no_write_without_approval, ...)

Export a JSON TrustOps policy for one MCP server with allow, blocked tools, required scopes, freshness, and approval gates.

入力スキーマ

{
  "type": "object",
  "properties": {
    "identifier": {
      "type": "string",
      "description": "Canonical server identifier in namespace/name format."
    },
    "client": {
      "type": "string"
    },
    "max_risk": {
      "type": "string",
      "enum": [
        "low",
        "medium",
        "high",
        "critical"
      ]
    },
    "requires_oauth": {
      "type": "boolean"
    },
    "no_write_without_approval": {
      "type": "boolean"
    },
    "max_freshness_hours": {
      "type": "number"
    }
  },
  "required": [
    "identifier"
  ]
}
🟢get_subscription_options

Return alert types, subscription channels, watch scopes, and existing subscription endpoints.

入力スキーマ

{
  "type": "object",
  "properties": {}
}
🟢get_gateway_options

Return Verify Gateway semantics, /v1/route versus /v1/decide guidance, mandate inputs, ledger evidence hooks, and decision fields.

入力スキーマ

{
  "type": "object",
  "properties": {}
}
🟢get_hosting_options

Return Verify Hosted MCP runtime/sandbox hosting capabilities, controls, and endpoints.

入力スキーマ

{
  "type": "object",
  "properties": {}
}
🟢list_agents(risk_level, limit)

Agent Sprawl Radar: list discovered AI agents, MCP servers, bots, workflows, risk levels, and owners.

入力スキーマ

{
  "type": "object",
  "properties": {
    "risk_level": {
      "type": "string",
      "enum": [
        "low",
        "medium",
        "high",
        "critical"
      ]
    },
    "limit": {
      "type": "integer",
      "minimum": 1,
      "maximum": 100
    }
  }
}
🟢get_agent(agent_id)

Agent Sprawl Radar: return one agent with tools, data access, findings, drift, and recommended action.

入力スキーマ

{
  "type": "object",
  "properties": {
    "agent_id": {
      "type": "string"
    }
  },
  "required": [
    "agent_id"
  ]
}
🟢score_agent(agent_id, agent)

Agent Sprawl Radar: score a normalized agent payload or an existing registry agent.

入力スキーマ

{
  "type": "object",
  "properties": {
    "agent_id": {
      "type": "string"
    },
    "agent": {
      "type": "object"
    }
  }
}
🟢explain_agent_risk(agent_id)

Agent Sprawl Radar: explain risk findings and the recommended action for one registry agent.

入力スキーマ

{
  "type": "object",
  "properties": {
    "agent_id": {
      "type": "string"
    }
  },
  "required": [
    "agent_id"
  ]
}
🟢list_high_risk_agents(limit)

Agent Sprawl Radar: list high and critical risk agents for review.

入力スキーマ

{
  "type": "object",
  "properties": {
    "limit": {
      "type": "integer",
      "minimum": 1,
      "maximum": 100
    }
  }
}
🟢get_drift_report(agent_id, limit)

Agent Sprawl Radar: return agent drift events across latest snapshots.

入力スキーマ

{
  "type": "object",
  "properties": {
    "agent_id": {
      "type": "string"
    },
    "limit": {
      "type": "integer",
      "minimum": 1,
      "maximum": 100
    }
  }
}
🟢generate_remediation_plan(agent_id)

Agent Sprawl Radar: generate deterministic remediation steps for risky or drifting agents.

入力スキーマ

{
  "type": "object",
  "properties": {
    "agent_id": {
      "type": "string"
    }
  },
  "required": [
    "agent_id"
  ]
}

推奨プロンプト

search_research
Search for information about [topic] using MCP Verify
想定されるツール: search
find_specific
Find [specific item] using MCP Verify
想定されるツール: search
retrieve_data
Get details about [item] from MCP Verify
想定されるツール: fetch
fetch_info
Fetch [information type] using MCP Verify
想定されるツール: fetch
list_items
List all [items] available in MCP Verify
想定されるツール: list_agents

コミュニティ

このサーバーを評価する

エビデンス

最近の観測

検証済みバージョンは記録されていませんツール 22 件
検証済みバージョンは記録されていませんツール 22 件