Watchdog for unattended AI agents: alerts, evidence checks and a verifiable proof per run.
Maintain this server? Claim it
Call this server's tools from your browser. Nothing to install.
RunVouch API key (free at runvouch.com)
Requests go through mcptry because browsers can't call most MCP servers directly. Header values are sent to the server for that request only and are never stored or logged.
Tool list from our last check. Connect live to refresh it.
Acknowledge an alert by id.
This tool isn't marked read-only, so it may change something on the server or in an account you connected.
Acknowledge an alert by id.
| Argument |
|---|
| Type |
|---|
| Description |
|---|
| alert_id* | integer |
{
"type": "object",
"required": [
"alert_id"
],
"properties": {
"alert_id": {
"type": "integer"
}
}
}Open (un-acknowledged) alerts: MISSED, FAILED, NO_EVIDENCE, BUDGET_RUN, BUDGET_DAY, RETRY_STORM, DRIFT, STALLED.
No arguments.
{
"type": "object",
"properties": {}
}Report that a run ended, with status and evidence dict (name -> bool or {type:'url',url}). Green run without evidence alerts.
| Argument | Type | Description |
|---|---|---|
| cost | number | |
| run_id* | string | |
| status | string | |
| tokens | integer | |
| evidence | object |
{
"type": "object",
"required": [
"run_id"
],
"properties": {
"cost": {
"type": "number"
},
"run_id": {
"type": "string"
},
"status": {
"enum": [
"ok",
"fail"
],
"type": "string"
},
"tokens": {
"type": "integer"
},
"evidence": {
"type": "object"
}
}
}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.
| Argument | Type | Description |
|---|---|---|
| run_id* | string |
{
"type": "object",
"required": [
"run_id"
],
"properties": {
"run_id": {
"type": "string"
}
}
}Report that a run of `agent` started. Returns run_id.
| Argument | Type | Description |
|---|---|---|
| agent* | string | |
| source | string |
{
"type": "object",
"required": [
"agent"
],
"properties": {
"agent": {
"type": "string"
},
"source": {
"type": "string"
}
}
}Recent runs of one agent.
| Argument | Type | Description |
|---|---|---|
| agent* | string | |
| limit | integer |
{
"type": "object",
"required": [
"agent"
],
"properties": {
"agent": {
"type": "string"
},
"limit": {
"type": "integer",
"default": 20
}
}
}Health of all watched agents: state (ok/alert/failed/unproven/running/waiting), last run, 24h cost, open alerts.
No arguments.
{
"type": "object",
"properties": {}
}