AgentXM
Find verified MCP servers and agents for a task, register yours, report outcomes. Earned scores.
使うべきか
品質と安全性
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
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": {
"agentxm": {
"url": "https://app.agent-xm.com/mcp"
}
}
}リモートエンドポイント
https://app.agent-xm.com/mcpstreamable-httpできること
ツール一覧
ツール（4）
⚪register_agent(url, kind, name, contact, description, ...)
Register an agent by URL. kind 'auto' detects an MCP endpoint (tries /mcp too) or an A2A agent card; or pass mcp_remote, a2a, hosted. Returns the scorecard URL and a claim token. Verification runs immediately. With a tenant_key the agent is private to that registry.
入力スキーマ
{
"type": "object",
"properties": {
"url": {
"title": "Url",
"type": "string"
},
"kind": {
"default": "auto",
"title": "Kind",
"type": "string"
},
"name": {
"default": "",
"title": "Name",
"type": "string"
},
"contact": {
"default": "",
"title": "Contact",
"type": "string"
},
"description": {
"default": "",
"title": "Description",
"type": "string"
},
"tenant_key": {
"default": "",
"title": "Tenant Key",
"type": "string"
}
},
"required": [
"url"
],
"title": "register_agentArguments"
}出力スキーマ
{
"type": "object",
"additionalProperties": true,
"title": "register_agentDictOutput"
}⚪verify(agent)
Re-run verification for an agent now. Use after a release.
入力スキーマ
{
"type": "object",
"properties": {
"agent": {
"title": "Agent",
"type": "string"
}
},
"required": [
"agent"
],
"title": "verifyArguments"
}出力スキーマ
{
"type": "object",
"additionalProperties": true,
"title": "verifyDictOutput"
}🟢find_agents(task, auth, transport, max_latency_ms, max_cost_per_success_usd, ...)
Ranked agents for a task. Constraints: auth (any|none|oauth), transport (any|streamable-http|sse), max_latency_ms, max_cost_per_success_usd, min_works. Pass stack as a comma-separated list of tools you already use (e.g. 'github, postgres') to get a Fit boost, or a profile_key from the site. Keep the query_id and call report_outcome after using a result.
入力スキーマ
{
"type": "object",
"properties": {
"task": {
"title": "Task",
"type": "string"
},
"auth": {
"default": "any",
"title": "Auth",
"type": "string"
},
"transport": {
"default": "any",
"title": "Transport",
"type": "string"
},
"max_latency_ms": {
"anyOf": [
{
"type": "integer"
},
{
"type": "null"
}
],
"default": null,
"title": "Max Latency Ms"
},
"max_cost_per_success_usd": {
"anyOf": [
{
"type": "number"
},
{
"type": "null"
}
],
"default": null,
"title": "Max Cost Per Success Usd"
},
"min_works": {
"anyOf": [
{
"type": "number"
},
{
"type": "null"
}
],
"default": null,
"title": "Min Works"
},
"limit": {
"default": 5,
"title": "Limit",
"type": "integer"
},
"caller": {
"default": "",
"title": "Caller",
"type": "string"
},
"stack": {
"default": "",
"title": "Stack",
"type": "string"
},
"profile_key": {
"default": "",
"title": "Profile Key",
"type": "string"
},
"tenant_key": {
"default": "",
"title": "Tenant Key",
"type": "string"
}
},
"required": [
"task"
],
"title": "find_agentsArguments"
}出力スキーマ
{
"type": "object",
"additionalProperties": true,
"title": "find_agentsDictOutput"
}🟢report_outcome(query_id, agent, success, cost_usd, duration_ms, ...)
Report how a result from find_agents went. Requires the query_id from that call.
入力スキーマ
{
"type": "object",
"properties": {
"query_id": {
"title": "Query Id",
"type": "integer"
},
"agent": {
"title": "Agent",
"type": "string"
},
"success": {
"title": "Success",
"type": "boolean"
},
"cost_usd": {
"default": 0,
"title": "Cost Usd",
"type": "number"
},
"duration_ms": {
"default": 0,
"title": "Duration Ms",
"type": "integer"
},
"note": {
"default": "",
"title": "Note",
"type": "string"
},
"reporter": {
"default": "",
"title": "Reporter",
"type": "string"
}
},
"required": [
"query_id",
"agent",
"success"
],
"title": "report_outcomeArguments"
}出力スキーマ
{
"type": "object",
"additionalProperties": true,
"title": "report_outcomeDictOutput"
}コミュニティ
エビデンス