Vivere
Know when cron jobs and AI agents stop running: create monitors and check in from your agent.
我該用這個嗎
品質與安全性
根據工具定義與協定合規性的自動化分析。
上下文成本
這是每次將伺服器的工具載入模型上下文時所消耗的約略 token 數量。數量越高，可用於其他工作的注意力就越少。
安裝
一鍵安裝
將以下內容加入你的 `claude_desktop_config.json` 檔案：
{
"mcpServers": {
"monitors": {
"url": "https://vivere.dev/mcp"
}
}
}遠端端點
https://vivere.dev/mcpstreamable-http它能做什麼
工具清單
工具（7）
🟢list_pages
List the pages this account can see: its own, then pages other accounts share with it, each with the access it has there (owner, editor or viewer) and how many monitors are on it and down.
輸入結構描述
{
"type": "object",
"properties": {}
}🟢list_monitors(page_id)
List the monitors on every page this account can see, or on one page, with status, page and access. Monitors on a page shared to view have no id; address them by page_id and slug.
輸入結構描述
{
"type": "object",
"properties": {
"page_id": {
"description": "Only this page (from list_pages)",
"type": "integer"
}
}
}🟢get_monitor(id, page_id, slug)
Get one monitor by id, including its ping URL for heartbeat monitors, or by page_id and slug, which also works on a page shared to view.
輸入結構描述
{
"type": "object",
"properties": {
"id": {
"description": "Monitor id, for monitors this account owns or edits",
"type": "string"
},
"page_id": {
"description": "Page id, with slug",
"type": "integer"
},
"slug": {
"description": "Monitor slug, with page_id",
"type": "string"
}
}
}🟡create_monitor(cron, grace, interval, kind, name, ...)
Create a heartbeat monitor (default) or an HTTP check on the account's first page or on page_id, which may be a page shared with this account as editor. Returns the monitor and, for heartbeats, the ping URL to call after each run.
輸入結構描述
{
"type": "object",
"properties": {
"cron": {
"description": "Heartbeat alternative to period: five-field cron expression, e.g. '0 2 * * *'.",
"type": "string"
},
"grace": {
"description": "Heartbeat: how late a check-in may be before alerting, e.g. 15m. Default 15m.",
"type": "string"
},
"interval": {
"description": "HTTP check: how often to request it, e.g. 5m.",
"type": "string"
},
"kind": {
"description": "heartbeat (the job pings us) or http (we request a URL)",
"enum": [
"heartbeat",
"http"
],
"type": "string"
},
"name": {
"description": "Human name, e.g. 'nightly report agent'",
"type": "string"
},
"page_id": {
"description": "Page to put it on (from list_pages). Default: the account's first page.",
"type": "integer"
},
"period": {
"description": "Heartbeat: expected time between successful check-ins, e.g. 30m, 1h, 1d. Default 1h.",
"type": "string"
},
"tags": {
"description": "Optional space-separated tags.",
"type": "string"
},
"timezone": {
"description": "IANA timezone for the cron expression, e.g. Asia/Seoul. Default UTC.",
"type": "string"
},
"url": {
"description": "HTTP check: the public URL to request.",
"type": "string"
}
},
"required": [
"name"
]
}🟢check_in(message, monitor_id, status)
Report on a heartbeat monitor: 'success' after a completed run (resets the deadline), 'start' when a run begins, 'fail' when it failed. Include a short message; it is kept as the run log.
輸入結構描述
{
"type": "object",
"properties": {
"message": {
"description": "Optional summary or error text to store with the run",
"type": "string"
},
"monitor_id": {
"description": "Monitor id (the last path segment of the ping URL)",
"type": "string"
},
"status": {
"description": "Default success",
"enum": [
"success",
"start",
"fail"
],
"type": "string"
}
},
"required": [
"monitor_id"
]
}⚪pause_monitor(id)
Pause a monitor so it stops alerting until resumed or pinged again.
輸入結構描述
{
"type": "object",
"properties": {
"id": {
"description": "Monitor id",
"type": "string"
}
},
"required": [
"id"
]
}⚪resume_monitor(id)
Resume a paused monitor.
輸入結構描述
{
"type": "object",
"properties": {
"id": {
"description": "Monitor id",
"type": "string"
}
},
"required": [
"id"
]
}社群
證據