Agent Radar
Search, inspect, recommend, and explain rated AI tools through Agent Radar.
我该使用它吗
质量与安全性
A
发现（3）
- LOWTool 'get_tool_card' description lacks action verb在 get_tool_card 中
- LOWTool 'recommend_tools' description lacks action verb在 recommend_tools 中
- LOWTool 'explain_rating' description lacks action verb在 explain_rating 中
基于对工具定义和协议合规性的自动分析。
上下文开销
~1,133token 数（工具定义）
~3.1 KB典型响应大小
对注意力有中等影响（占 128k 上下文窗口的 0.89%）
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": {
"agent-radar": {
"url": "https://agent-radar.zation1.workers.dev/api/mcp"
}
}
}远程端点
https://agent-radar.zation1.workers.dev/api/mcpstreamable-http它能做什么
工具清单
工具（4）
🟢 只读🟡 写入🔴 删除⚪ 未知
🟢search_tools(query, top_k, filters)
Search Agent Radar Tool Cards by query text and optional type, tag, or risk filters.
输入模式
{
"type": "object",
"properties": {
"query": {
"default": "",
"description": "Free-text tool name, capability, or task to search for. Use an empty string to browse the highest-rated matching catalog entries.",
"type": "string"
},
"top_k": {
"default": 5,
"description": "Maximum number of results to return, from 1 to 50. Defaults to 5.",
"type": "integer",
"minimum": 1,
"maximum": 50
},
"filters": {
"description": "Optional exact-match filters applied before search ranking.",
"type": "object",
"properties": {
"type": {
"description": "Exact Agent Radar tool type to include, such as mcp, skill, agent, framework, or cli.",
"type": "string",
"enum": [
"mcp",
"skill",
"agent",
"framework",
"cli",
"prompt",
"rules",
"dataset",
"service"
]
},
"tags": {
"description": "Exact catalog tags that every returned tool must contain.",
"type": "array",
"items": {
"type": "string"
}
},
"risk_level": {
"description": "Exact assessed risk level to include: low, medium, high, critical, or unknown.",
"type": "string",
"enum": [
"low",
"medium",
"high",
"critical",
"unknown"
]
}
},
"additionalProperties": false
}
},
"$schema": "https://json-schema.org/draft/2020-12/schema",
"additionalProperties": false
}输出模式
{
"type": "object",
"properties": {},
"$schema": "https://json-schema.org/draft/2020-12/schema",
"additionalProperties": {}
}🟢get_tool_card(tool_id)
Return one Tool Card and its Rating Result by stable tool_id.
输入模式
{
"type": "object",
"properties": {
"tool_id": {
"type": "string",
"minLength": 1,
"description": "Stable Agent Radar tool identifier, usually obtained from search_tools results."
}
},
"required": [
"tool_id"
],
"$schema": "https://json-schema.org/draft/2020-12/schema",
"additionalProperties": false
}输出模式
{
"type": "object",
"properties": {},
"$schema": "https://json-schema.org/draft/2020-12/schema",
"additionalProperties": {}
}🟢recommend_tools(task, language_or_stack, environment, preferred_tool_types, allowed_permissions, ...)
Recommend known tools for a task using request-scoped LLM credentials when required.
输入模式
{
"type": "object",
"properties": {
"task": {
"type": "string",
"minLength": 1,
"description": "Natural-language development task for which Agent Radar should recommend suitable AI tools."
},
"language_or_stack": {
"description": "Programming languages, frameworks, runtimes, or other stack constraints relevant to the task.",
"type": "array",
"items": {
"type": "string"
}
},
"environment": {
"description": "Execution contexts such as local development, CI, browser, cloud, IDE, or production.",
"type": "array",
"items": {
"type": "string"
}
},
"preferred_tool_types": {
"description": "Agent Radar tool types to prefer when selecting candidates.",
"type": "array",
"items": {
"type": "string",
"enum": [
"mcp",
"skill",
"agent",
"framework",
"cli",
"prompt",
"rules",
"dataset",
"service"
]
}
},
"allowed_permissions": {
"description": "Permission scope names the caller is willing to allow; candidates outside these boundaries are treated conservatively.",
"type": "array",
"items": {
"type": "string"
}
},
"risk_tolerance": {
"description": "Maximum preferred risk tolerance for the recommendation: low, medium, or high.",
"type": "string",
"enum": [
"low",
"medium",
"high"
]
},
"existing_tools": {
"description": "Tools already available to the project or agent, used as compatibility and duplication context.",
"type": "array",
"items": {
"type": "string"
}
},
"budget": {
"description": "Natural-language cost constraint, such as free, free_or_low_cost, or a project-specific budget.",
"type": "string"
},
"output_format": {
"description": "Preferred presentation format for recommendation content: json or markdown. MCP still returns structured content.",
"type": "string",
"enum": [
"json",
"markdown"
]
},
"top_k": {
"description": "Maximum number of recommended candidates to return, from 1 to 50. Defaults to 5 when omitted.",
"type": "integer",
"minimum": 1,
"maximum": 50
},
"model": {
"description": "LLM provider model identifier recognized by Agent Radar. Omit it to use the server-configured or registry default model.",
"type": "string",
"minLength": 1
}
},
"required": [
"task"
],
"$schema": "https://json-schema.org/draft/2020-12/schema",
"additionalProperties": false
}输出模式
{
"type": "object",
"properties": {},
"$schema": "https://json-schema.org/draft/2020-12/schema",
"additionalProperties": {}
}🟢explain_rating(tool_id)
Return rating dimensions, explanations, penalties, and boosts for one tool_id.
输入模式
{
"type": "object",
"properties": {
"tool_id": {
"type": "string",
"minLength": 1,
"description": "Stable Agent Radar tool identifier, usually obtained from search_tools results."
}
},
"required": [
"tool_id"
],
"$schema": "https://json-schema.org/draft/2020-12/schema",
"additionalProperties": false
}输出模式
{
"type": "object",
"properties": {},
"$schema": "https://json-schema.org/draft/2020-12/schema",
"additionalProperties": {}
}社区
证据
最近观测
已验证未记录版本4 个工具
已验证未记录版本4 个工具