runvouch
Watchdog for unattended AI agents: alerts, evidence checks and a verifiable proof per run.
我该使用它吗
质量与安全性
发现（2）
- HIGHTool poisoning patterns detected
- MEDIUMTool description contains suspicious base64-like encoded string在 runvouch_status 中
基于对工具定义和协议合规性的自动分析。
上下文开销
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": {
"runvouch": {
"url": "https://api.runvouch.com/mcp"
}
}
}远程端点
https://api.runvouch.com/mcpstreamable-http它能做什么
工具清单
工具（7）
🟢runvouch_status
Health of all watched agents: state (ok/alert/failed/unproven/running/waiting), last run, 24h cost, open alerts.
输入模式
{
"type": "object",
"properties": {}
}🟢runvouch_alerts
Open (un-acknowledged) alerts: MISSED, FAILED, NO_EVIDENCE, BUDGET_RUN, BUDGET_DAY, RETRY_STORM, DRIFT, STALLED.
输入模式
{
"type": "object",
"properties": {}
}⚪runvouch_ack(alert_id)
Acknowledge an alert by id.
输入模式
{
"type": "object",
"properties": {
"alert_id": {
"type": "integer"
}
},
"required": [
"alert_id"
]
}🟢runvouch_runs(agent, limit)
Recent runs of one agent.
输入模式
{
"type": "object",
"properties": {
"agent": {
"type": "string"
},
"limit": {
"type": "integer",
"default": 20
}
},
"required": [
"agent"
]
}⚪runvouch_run_start(agent, source)
Report that a run of `agent` started. Returns run_id.
输入模式
{
"type": "object",
"properties": {
"agent": {
"type": "string"
},
"source": {
"type": "string"
}
},
"required": [
"agent"
]
}⚪runvouch_run_end(run_id, status, cost, tokens, evidence)
Report that a run ended, with status and evidence dict (name -> bool or {type:'url',url}). Green run without evidence alerts.
输入模式
{
"type": "object",
"properties": {
"run_id": {
"type": "string"
},
"status": {
"type": "string",
"enum": [
"ok",
"fail"
]
},
"cost": {
"type": "number"
},
"tokens": {
"type": "integer"
},
"evidence": {
"type": "object"
}
},
"required": [
"run_id"
]
}🟢runvouch_run_proof(run_id)
Tamper-evident proof of a finished run: hashed record, Merkle path, day root, chain hash and OpenTimestamps status. Verify offline with templates/verify_proof.py.
输入模式
{
"type": "object",
"properties": {
"run_id": {
"type": "string"
}
},
"required": [
"run_id"
]
}社区
证据