A live demo with sample data: look around, nothing can be changed. Start a free trial or self-host it.

search_hotels

Last 30 days

Calls

0

Nothing yet in this window

Error rate

0%

0 failed calls

Response time

-

median

Called from

-

Calls over time

CallsErrors
Show the numbers
Calls and errors per bucket
TimeCallsErrors
00
00
00
00
00
00
00
00
00
00
00
00
00
00
00
00
00
00
00
00
00
00
00
00
00
00
00
00
00
00
00

Response times

Calls by how long they took

No calls in this window.

Right before it went wrong

Before 17 repeated or failed calls, in their sessions

What the agent called just before search_hotels repeated itself or failed. One call standing out usually did not do what the agent expected: a click that did not land, a search that came back empty.

Called beforeClientRepeatedFailed
get_flight_statusClaude02
Prompt plan_tripClaude Code02
Prompt summarize_bookingClaude Code02
Resource airports://{code}Claude01
Resource airports://{code}Other01
get_flight_statusChatGPT01
Prompt plan_tripChatGPT01
Prompt plan_tripClaude01
Prompt plan_tripCursor01
search_flightsClaude Code01
Prompt summarize_bookingOther01
Resource trips://{id}Claude01
Resource trips://{id}Cursor01
Resource trips://{id}MCP Inspector01

How it fails

Nothing failed in this window.

Called from

Not called in this window.

Parameters agents send

Names and types only

Not called in this window.