Skip to main content

Troubleshoot by failure stage

Find the responsible boundary before changing prompts, providers, or infrastructure.
2 min read

Identify the failing stage#

Record the workspace, agent ID, run ID when available, time, call direction, channel, and exact error. Remove secrets before sharing logs.

SymptomCheck firstReference
API returns 401API key, header, and owning organizationAuthentication
API returns 402Feature entitlement, workflow limit, or inference balance named in the errorBilling
Draft will not publishValidation errors, complete graph, and conflicting triggersVersions
Call starts without audioMicrophone, model validation, WebRTC/ICE, and TURNSetup troubleshooting
Inbound number reaches the wrong agentNumber mapping, provider webhook, and published triggerInbound routing
Outbound initiation succeeds but no conversation occursCarrier events, destination restrictions, caller ID, and mediaTelephony
Tool is never calledNode attachment, prompt instruction, schema, and required valuesTools
Knowledge produces no answerDocument processing state, embedding configuration, and query relevanceKnowledge base
Webhook did not update your systemCompleted run, template values, credentials, receiver logs, and retry policyWebhook payloads
Production differs from an editor testPublished definition, runtime context, and model overridesTesting

Change one responsible part#

A healthy homepage does not establish that model calls, media, storage, or a queue worker work. Reproduce the smallest failing operation, inspect its evidence, fix that boundary, and run the same check again.

Escalate with useful evidence#

Provide the exact failing step, expected and actual result, timestamp, relevant IDs, and redacted error text. Include whether text chat, Web Call, telephony, and downstream delivery passed independently. For self-hosting, include the deployed image or commit and the affected service logs.