Call this server's tools from your browser. Nothing to install.
Bearer dm_… — an account token from Account → API access at https://dealmonitor.app. Leave it out and price checks still work; anything that keeps a watch needs it.
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.
Read the current price of a product from its shop URL. Works on shops that block simple HTTP requests, because the page is fetched server-side with a real browser when needed. No account required. Can take up to two minutes on a slow shop.
This tool isn't marked read-only, so it may change something on the server or in an account you connected.
Pick your client. Values in capitals, like YOUR_TOKEN, are yours to fill in.
Run in a terminal
claude mcp add --transport http dealmonitor https://mcp.dealmonitor.app/mcp
Headers
Authorization (optional, not in the config above): Bearer dm_… — an account token from Account → API access at https://dealmonitor.app. Leave it out and price checks still work; anything that keeps a watch needs it.Read the current price of a product from its shop URL. Works on shops that block simple HTTP requests, because the page is fetched server-side with a real browser when needed. No account required. Can take up to two minutes on a slow shop.
| Argument | Type | Description |
|---|---|---|
| url* | string |
{
"type": "object",
"title": "check_priceArguments",
"required": [
"url"
],
"properties": {
"url": {
"type": "string",
"title": "Url"
}
}
}List the price watches on this account, with each one's current price, target and whether the target is met right now. Use only_reached=true to get just the ones worth acting on.
| Argument | Type | Description |
|---|---|---|
| only_reached | boolean |
{
"type": "object",
"title": "list_watchesArguments",
"properties": {
"only_reached": {
"type": "boolean",
"title": "Only Reached",
"default": false
}
}
}Price history for one watch, oldest first. Use it to say whether the current price is actually a good one rather than only below a target.
| Argument | Type | Description |
|---|---|---|
| limit | integer | |
| watch_id* | integer |
{
"type": "object",
"title": "price_historyArguments",
"required": [
"watch_id"
],
"properties": {
"limit": {
"type": "integer",
"title": "Limit",
"default": 60
},
"watch_id": {
"type": "integer",
"title": "Watch Id"
}
}
}Change the target price of an existing watch.
| Argument | Type | Description |
|---|---|---|
| watch_id* | integer | |
| target_price* | number |
{
"type": "object",
"title": "set_target_priceArguments",
"required": [
"watch_id",
"target_price"
],
"properties": {
"watch_id": {
"type": "integer",
"title": "Watch Id"
},
"target_price": {
"type": "number",
"title": "Target Price"
}
}
}Stop watching a product and delete the watch, including its recorded price history for this account. Not reversible.
| Argument | Type | Description |
|---|---|---|
| watch_id* | integer |
{
"type": "object",
"title": "stop_watchingArguments",
"required": [
"watch_id"
],
"properties": {
"watch_id": {
"type": "integer",
"title": "Watch Id"
}
}
}Set a price alert: keep watching a product's price and notify the account owner when it reaches the target. Use this for any 'tell me when this drops', 'notify me at X', 'monitor this price' or 'keep an eye on this' task. Reads the current price in the same call, so one call answers both 'what does it cost now' and 'tell me when it gets cheaper'. The watch keeps running after this conversation ends. Requires an API token.
| Argument | Type | Description |
|---|---|---|
| url* | string | |
| target_price* | number |
{
"type": "object",
"title": "watch_priceArguments",
"required": [
"url",
"target_price"
],
"properties": {
"url": {
"type": "string",
"title": "Url"
},
"target_price": {
"type": "number",
"title": "Target Price"
}
}
}We read every tool's name, description and inputs the way the model will, looking for instructions hidden from you. A clean result means no warning signs, not a guarantee. Scanned Nov 21, 2026.
No readOnlyHint / destructiveHint annotations, so clients can't tell a read from a write without asking.
Fix: Add annotations to each tool, e.g. { readOnlyHint: true } for tools that only read.