MCP MeasureMCP Measure

索引已观测

SYMBaiEX Evidence MCP

Authenticated public evidence search, verification, research jobs, exports, and webhooks.

我该使用它吗

质量与安全性

B
描述质量
94%
模式完整度
69%
命名质量
79%
投毒风险
100%
权限匹配度
30%
协议合规性
100%

发现（9）

  • LOWTool 'evidence_rotate_webhook_secret' description lacks action verb在 evidence_rotate_webhook_secret 中
  • LOWTool 'evidence_replay_webhook_delivery' name length outside 3-30 range在 evidence_replay_webhook_delivery 中
  • LOWTool 'evidence_list_webhooks' suggests web access but openWorldHint=false在 evidence_list_webhooks 中
  • LOWTool 'evidence_webhook_availability' suggests web access but openWorldHint=false在 evidence_webhook_availability 中
  • LOWTool 'evidence_create_webhook' suggests web access but openWorldHint=false在 evidence_create_webhook 中
  • LOWTool 'evidence_set_webhook_status' suggests web access but openWorldHint=false在 evidence_set_webhook_status 中
  • LOWTool 'evidence_webhook_health' suggests web access but openWorldHint=false在 evidence_webhook_health 中
  • LOWTool 'evidence_webhook_history' suggests web access but openWorldHint=false在 evidence_webhook_history 中
  • LOWTool 'evidence_rotate_webhook_secret' suggests web access but openWorldHint=false在 evidence_rotate_webhook_secret 中

基于对工具定义和协议合规性的自动分析。

上下文开销

~3,303token 数（工具定义）
~552 B典型响应大小
对注意力有显著影响（占 128k 上下文窗口的 2.58%）

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": {
    "evidence": {
      "url": "https://www.symbaiex.com/api/agent/mcp"
    }
  }
}

远程端点

https://www.symbaiex.com/api/agent/mcpstreamable-http

它能做什么

工具清单

工具（34）

🟢 只读🟡 写入🔴 删除⚪ 未知
🟢evidence_status

Read product API availability and capability status.

输入模式

{
  "type": "object",
  "properties": {},
  "required": [],
  "additionalProperties": false
}
🟢evidence_sources(limit)

List current active public evidence sources.

输入模式

{
  "type": "object",
  "properties": {
    "limit": {
      "type": "integer",
      "minimum": 1,
      "maximum": 50,
      "default": 20
    }
  },
  "required": [],
  "additionalProperties": false
}
🟢evidence_source_registry

Read deterministic adapter provenance, cadence, and ingestion bounds.

输入模式

{
  "type": "object",
  "properties": {},
  "required": [],
  "additionalProperties": false
}
🟢evidence_search(query, limit)

Search normalized current public evidence documents.

输入模式

{
  "type": "object",
  "properties": {
    "query": {
      "type": "string",
      "minLength": 1,
      "maxLength": 300
    },
    "limit": {
      "type": "integer",
      "minimum": 1,
      "maximum": 50,
      "default": 20
    }
  },
  "required": [
    "query"
  ],
  "additionalProperties": false
}
🟢evidence_semantic_search(query, limit, idempotencyKey)

Search the public corpus with feature-gated semantic enhancement and lexical fallback.

输入模式

{
  "type": "object",
  "properties": {
    "query": {
      "type": "string",
      "minLength": 1,
      "maxLength": 300
    },
    "limit": {
      "type": "integer",
      "minimum": 1,
      "maximum": 20,
      "default": 10
    },
    "idempotencyKey": {
      "type": "string",
      "minLength": 8,
      "maxLength": 96,
      "pattern": "^[A-Za-z0-9][A-Za-z0-9._:-]*$"
    }
  },
  "required": [
    "query",
    "idempotencyKey"
  ],
  "additionalProperties": false
}
🟢evidence_universal_search(query, limit)

Search public evidence, editorial, news, newsletter, and forum records.

输入模式

{
  "type": "object",
  "properties": {
    "query": {
      "type": "string",
      "minLength": 1,
      "maxLength": 300
    },
    "limit": {
      "type": "integer",
      "minimum": 1,
      "maximum": 50,
      "default": 20
    }
  },
  "required": [
    "query"
  ],
  "additionalProperties": false
}
🟢evidence_document(documentId)

Read one public evidence document by its ID.

输入模式

{
  "type": "object",
  "properties": {
    "documentId": {
      "type": "string",
      "minLength": 1,
      "maxLength": 128
    }
  },
  "required": [
    "documentId"
  ],
  "additionalProperties": false
}
🟢evidence_changes(limit)

List recent public evidence change events.

输入模式

{
  "type": "object",
  "properties": {
    "limit": {
      "type": "integer",
      "minimum": 1,
      "maximum": 50,
      "default": 20
    }
  },
  "required": [],
  "additionalProperties": false
}
⚪evidence_verify_claim(claimId)

Verify one stored public claim against version-bound citations.

输入模式

{
  "type": "object",
  "properties": {
    "claimId": {
      "type": "string",
      "minLength": 1,
      "maxLength": 128
    }
  },
  "required": [
    "claimId"
  ],
  "additionalProperties": false
}
🟢evidence_search_entities(query, limit)

Search normalized public evidence entities.

输入模式

{
  "type": "object",
  "properties": {
    "query": {
      "type": "string",
      "minLength": 1,
      "maxLength": 200
    },
    "limit": {
      "type": "integer",
      "minimum": 1,
      "maximum": 50,
      "default": 20
    }
  },
  "required": [
    "query"
  ],
  "additionalProperties": false
}
🟢evidence_version_claims(documentVersionId, limit)

List public claims cited to one evidence document version.

输入模式

{
  "type": "object",
  "properties": {
    "documentVersionId": {
      "type": "string",
      "minLength": 1,
      "maxLength": 128
    },
    "limit": {
      "type": "integer",
      "minimum": 1,
      "maximum": 50,
      "default": 20
    }
  },
  "required": [
    "documentVersionId"
  ],
  "additionalProperties": false
}
🟢evidence_claim_timeline(claimKey, limit)

Read the bounded public supersession timeline for a claim key.

输入模式

{
  "type": "object",
  "properties": {
    "claimKey": {
      "type": "string",
      "minLength": 1,
      "maxLength": 200
    },
    "limit": {
      "type": "integer",
      "minimum": 1,
      "maximum": 50,
      "default": 20
    }
  },
  "required": [
    "claimKey"
  ],
  "additionalProperties": false
}
🟢evidence_products(limit)

List current evidence API product definitions.

输入模式

{
  "type": "object",
  "properties": {
    "limit": {
      "type": "integer",
      "minimum": 1,
      "maximum": 50,
      "default": 20
    }
  },
  "required": [],
  "additionalProperties": false
}
🟢evidence_datasets(limit)

List public dataset selection contracts and immutable snapshot provenance, coverage, freshness, and citation metrics.

输入模式

{
  "type": "object",
  "properties": {
    "limit": {
      "type": "integer",
      "minimum": 1,
      "maximum": 50,
      "default": 20
    }
  },
  "required": [],
  "additionalProperties": false
}
🟢evidence_benchmarks(limit)

List public benchmark definitions and recent runs.

输入模式

{
  "type": "object",
  "properties": {
    "limit": {
      "type": "integer",
      "minimum": 1,
      "maximum": 50,
      "default": 20
    }
  },
  "required": [],
  "additionalProperties": false
}
🟢evidence_quota

Read the authenticated agent's current usage quota.

输入模式

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

Read effective free-beta product access for the authenticated owner.

输入模式

{
  "type": "object",
  "properties": {},
  "required": [],
  "additionalProperties": false
}
🟢evidence_usage(limit)

Read bounded authenticated usage ledger summaries.

输入模式

{
  "type": "object",
  "properties": {
    "limit": {
      "type": "integer",
      "minimum": 1,
      "maximum": 50,
      "default": 20
    }
  },
  "required": [],
  "additionalProperties": false
}
🟡evidence_create_research(query)

Create one bounded source-cited research job.

输入模式

{
  "type": "object",
  "properties": {
    "query": {
      "type": "string",
      "minLength": 1,
      "maxLength": 300
    }
  },
  "required": [
    "query"
  ],
  "additionalProperties": false
}
🟢evidence_get_research(jobId)

Read one owner-scoped research job status.

输入模式

{
  "type": "object",
  "properties": {
    "jobId": {
      "type": "string",
      "minLength": 1,
      "maxLength": 128
    }
  },
  "required": [
    "jobId"
  ],
  "additionalProperties": false
}
🟢evidence_get_research_artifact(jobId)

Resolve one authenticated first-party research artifact stream.

输入模式

{
  "type": "object",
  "properties": {
    "jobId": {
      "type": "string",
      "minLength": 1,
      "maxLength": 128
    }
  },
  "required": [
    "jobId"
  ],
  "additionalProperties": false
}
🟡evidence_create_export(datasetId, format, recordLimit, selectionStrategy, cursor)

Create one bounded public JSONL export job.

输入模式

{
  "type": "object",
  "properties": {
    "datasetId": {
      "type": "string",
      "minLength": 1,
      "maxLength": 128
    },
    "format": {
      "type": "string",
      "enum": [
        "jsonl",
        "parquet"
      ],
      "default": "jsonl"
    },
    "recordLimit": {
      "type": "integer",
      "minimum": 1,
      "maximum": 500,
      "default": 25
    },
    "selectionStrategy": {
      "type": "string",
      "enum": [
        "balanced",
        "recent"
      ],
      "default": "balanced"
    },
    "cursor": {
      "type": "string",
      "minLength": 1,
      "maxLength": 512
    }
  },
  "required": [
    "datasetId"
  ],
  "additionalProperties": false
}
🟢evidence_get_export(jobId)

Read one owner-scoped export job status.

输入模式

{
  "type": "object",
  "properties": {
    "jobId": {
      "type": "string",
      "minLength": 1,
      "maxLength": 128
    }
  },
  "required": [
    "jobId"
  ],
  "additionalProperties": false
}
🟢evidence_get_export_artifact(jobId)

Resolve one authenticated first-party export artifact stream.

输入模式

{
  "type": "object",
  "properties": {
    "jobId": {
      "type": "string",
      "minLength": 1,
      "maxLength": 128
    }
  },
  "required": [
    "jobId"
  ],
  "additionalProperties": false
}
🟢evidence_get_export_manifest(jobId)

Resolve the immutable buyer-readable provenance and coverage manifest for an export.

输入模式

{
  "type": "object",
  "properties": {
    "jobId": {
      "type": "string",
      "minLength": 1,
      "maxLength": 128
    }
  },
  "required": [
    "jobId"
  ],
  "additionalProperties": false
}
🟢evidence_list_webhooks

List owner-scoped webhook subscriptions.

输入模式

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

Read whether webhook creation is currently operator-enabled.

输入模式

{
  "type": "object",
  "properties": {},
  "required": [],
  "additionalProperties": false
}
🟡evidence_create_webhook(subscriptionKey, url, eventTypes)

Create an owner-scoped allowlisted signed webhook subscription.

输入模式

{
  "type": "object",
  "properties": {
    "subscriptionKey": {
      "type": "string",
      "minLength": 1,
      "maxLength": 96
    },
    "url": {
      "type": "string",
      "format": "uri",
      "maxLength": 2000
    },
    "eventTypes": {
      "type": "array",
      "minItems": 1,
      "maxItems": 12,
      "uniqueItems": true,
      "items": {
        "type": "string",
        "enum": [
          "evidence.document.changed",
          "evidence.research.completed",
          "evidence.export.completed",
          "evidence.benchmark.completed",
          "evidence.claim.verified",
          "evidence.usage.threshold_reached",
          "evidence.usage.budget_reached",
          "evidence.entitlement.status_changed"
        ]
      }
    }
  },
  "required": [
    "subscriptionKey",
    "url",
    "eventTypes"
  ],
  "additionalProperties": false
}
🟡evidence_set_webhook_status(subscriptionKey, status)

Pause, resume, or revoke an owner-scoped webhook subscription.

输入模式

{
  "type": "object",
  "properties": {
    "subscriptionKey": {
      "type": "string",
      "minLength": 1,
      "maxLength": 96
    },
    "status": {
      "type": "string",
      "enum": [
        "active",
        "paused",
        "revoked"
      ]
    }
  },
  "required": [
    "subscriptionKey",
    "status"
  ],
  "additionalProperties": false
}
🟢evidence_webhook_health(subscriptionKey)

Read bounded health for one owner-scoped webhook subscription.

输入模式

{
  "type": "object",
  "properties": {
    "subscriptionKey": {
      "type": "string",
      "minLength": 1,
      "maxLength": 96
    }
  },
  "required": [
    "subscriptionKey"
  ],
  "additionalProperties": false
}
🟢evidence_webhook_history(subscriptionKey, limit)

List bounded recent deliveries for one webhook subscription.

输入模式

{
  "type": "object",
  "properties": {
    "subscriptionKey": {
      "type": "string",
      "minLength": 1,
      "maxLength": 96
    },
    "limit": {
      "type": "integer",
      "minimum": 1,
      "maximum": 50,
      "default": 20
    }
  },
  "required": [
    "subscriptionKey"
  ],
  "additionalProperties": false
}
⚪evidence_rotate_webhook_secret(subscriptionKey)

Rotate a webhook signing secret and return it once.

输入模式

{
  "type": "object",
  "properties": {
    "subscriptionKey": {
      "type": "string",
      "minLength": 1,
      "maxLength": 96
    }
  },
  "required": [
    "subscriptionKey"
  ],
  "additionalProperties": false
}
⚪evidence_replay_webhook_delivery(subscriptionKey, eventKey)

Replay one stored webhook event with bounded replay protection.

输入模式

{
  "type": "object",
  "properties": {
    "subscriptionKey": {
      "type": "string",
      "minLength": 1,
      "maxLength": 96
    },
    "eventKey": {
      "type": "string",
      "minLength": 1,
      "maxLength": 160
    }
  },
  "required": [
    "subscriptionKey",
    "eventKey"
  ],
  "additionalProperties": false
}
🟡evidence_send_webhook_test(subscriptionKey, eventType)

Send one signed test event to an active webhook subscription.

输入模式

{
  "type": "object",
  "properties": {
    "subscriptionKey": {
      "type": "string",
      "minLength": 1,
      "maxLength": 96
    },
    "eventType": {
      "type": "string",
      "enum": [
        "evidence.document.changed",
        "evidence.research.completed",
        "evidence.export.completed",
        "evidence.benchmark.completed",
        "evidence.claim.verified",
        "evidence.usage.threshold_reached",
        "evidence.usage.budget_reached",
        "evidence.entitlement.status_changed"
      ]
    }
  },
  "required": [
    "subscriptionKey",
    "eventType"
  ],
  "additionalProperties": false
}

社区

评价此服务器

证据

最近观测

已验证未记录版本34 个工具
已验证未记录版本34 个工具
已验证未记录版本34 个工具