Tool health
See how each tool your agents use behaved in real conversations and calls: runs, failures, empty results, platform blocks, top errors, and examples.
Before you start
- MCP keys need both
workflows:readandleads:read, because examples include lead data. Full-access REST keys need nothing else. - Test leads, sandbox leads, and Playground conversations are not counted.
- The report is read-only. Calling it does not run any tool.
- Errors return a stable code in
error, such asinvalid_days. Map the code to your own copy.
Choose what to read
What each number means
How to read the results
Limits
- At most 100 rows.
truncatedistruewhen more matched. - One request reads the most recent activity first. When the window had more tool activity than one request reads,
truncatedistrueandwindow.scanned_fromshows the moment the counts start from. Narrow withworkflow_id,tool_name,channel, or fewerdays. argsandresultare cut at 600 characters. API keys, tokens, passwords, and signatures are masked. Lead data such as names and phone numbers is not removed, so handle samples as lead data.
Authorizations
API key for authentication. Keys are prefixed with nxr_live_.
Query Parameters
How many days back to read, counted in 24-hour periods from the moment of the request. Any other value returns 400 invalid_days.
1 <= x <= 14Task ID (workflow in the API). Counts only that task's conversations and calls. A malformed ID returns 400 invalid_workflow_id; a task from another account returns an empty tools list.
Exact tool name. Surrounding spaces are ignored. Empty after trimming, or longer than 120 characters, returns 400 invalid_tool_name.
120Only runs on this channel. call is phone calls; unsupervised-whatsapp is a personal WhatsApp line. Any other value returns 400 invalid_channel.
whatsapp, unsupervised-whatsapp, instagram, messenger, webchat, sms, imessage, email, call true keeps only tools where failed + empty is at least half of the runs that reached the integration (calls - blocked), with at least 3 such runs. Accepts true, false, 1, or 0; anything else returns 400 invalid_only_problems.
Response
Tool rows, sorted by problem rate ((failed + empty) / (calls - blocked)) and then by calls, highest first.
100true when more than 100 rows matched, or when the window had more activity than one request reads (see window.scanned_from). Narrow with workflow_id, tool_name, channel, or fewer days.