Chat Test Results with Real JWT Token

Date: December 31, 2025
Test Method: Command-line with authenticated JWT token
Status: ❌ Agent execution failing (authentication ✅ working)

Test Results

Request

  • URL: https://api.engram.work/api/v1/chat
  • Method: POST
  • Authentication: ✅ Valid JWT token (Bearer token)
  • Payload: 
    {
  "content": "hi",
  "agent_id": "elena"
}

Response

  • Status: 200 OK ✅
  • Response Time: ~1522ms
  • Authentication: ✅ WORKING (token accepted)
  • Agent Execution: ❌ FAILING (error message returned)
  • LLM Call: ❌ FAILING (tokens_used: 0)

Response Body 

{
  "message_id": "584d1391-e92a-426b-87b7-1b393de0cf0d",
  "content": "I apologize, but I encountered an issue processing your request. Could you please try again? If the problem persists, the team can check the logs for more details.",
  "agent_id": "elena",
  "agent_name": "Dr. Elena Vasquez",
  "timestamp": "2025-12-31T23:28:11.957420Z",
  "tokens_used": 0,
  "latency_ms": 1521.46315574646,
  "session_id": "5d8965ce-712b-4b0d-9e5e-4e51809d3cf7"
}

Key Findings

✅ What’s Working

  1. API Endpoint: Responding correctly (200 OK)
  2. Authentication: ✅ JWT token validation is working correctly
  3. Request Routing: Request reaches chat handler
  4. Error Handling: Exception is caught and handled gracefully

❌ What’s Failing

  1. Agent Execution: The agent_chat() function is throwing an exception
  2. LLM API Call: No tokens consumed (tokens_used: 0) = LLM call failed
  3. Response Generation: Falling back to generic error message

Root Cause

The tokens_used: 0 is the definitive indicator that the LLM API call is failing. The authentication is working, so the issue is NOT with user authentication.

Most Likely Causes

1. LLM API Configuration Issue (Most Likely)

The Azure AI Foundry API call is failing. Possible causes:

  • Endpoint format: AZURE_AI_ENDPOINT missing /openai/v1/
  • API version: AZURE_AI_API_VERSION not 2024-12-01-preview
  • Model Router: AZURE_AI_MODEL_ROUTER still enabled (should be empty)
  • Deployment name: AZURE_AI_DEPLOYMENT not gpt-5.1-chat
  • API key: AZURE_AI_KEY missing or invalid

2. LLM API 400 Bad Request

Azure AI Foundry is returning a 400 error, likely due to:

  • API version mismatch
  • Model name mismatch
  • Endpoint format issue
  • Invalid request payload

3. Network/Connectivity Issue

The backend cannot reach the Azure AI Foundry endpoint (less likely, as this would show connection timeout).

Next Steps: Check Backend Logs

The improved logging should show the exact error. Check Azure Container Apps logs:

# View recent logs
az containerapp logs show \
  --name staging-env-api \
  --resource-group zimax-ai \
  --tail 200 \
  --follow

What to Look For

The logs should contain these messages (from the improved logging):

  1. Agent execution start: 
    INFO: Calling agent_chat for user d240186f-..., session ..., agent elena
  2. LLM API call attempt: 
    INFO: FoundryChatClient: Calling https://zimax-gw.azure-api.net/zimax/openai/v1/chat/completions
INFO: FoundryChatClient: is_openai_compat=True, model=gpt-5.1-chat
  3. Error messages: 
    ERROR: Agent execution failed: {error}
ERROR: FoundryChatClient: Error calling LLM: {error}
ERROR: FoundryChatClient: Error response body: {error_body}
ERROR: Full traceback: {traceback}

Expected Error Patterns

If LLM API returns 400:

ERROR: FoundryChatClient: Response status=400
ERROR: FoundryChatClient: Error response body: {"error": {"message": "Model o4-mini is enabled only for api versions 2024-12-01-preview and later"}}

If endpoint format is wrong:

ERROR: FoundryChatClient: Error calling LLM: 404 Not Found
ERROR: FoundryChatClient: Request URL: https://zimax-gw.azure-api.net/zimax/openai/deployments/gpt-5.1-chat/chat/completions

If API key is missing:

ERROR: FoundryChatClient: Error calling LLM: 401 Unauthorized

Verify Azure Configuration

Check environment variables in Azure Container Apps:

az containerapp show \
  --name staging-env-api \
  --resource-group zimax-ai \
  --query "properties.template.containers[0].env" \
  --output table

Required values:

  • AZURE_AI_ENDPOINT = https://zimax-gw.azure-api.net/zimax/openai/v1/ ✅ (must include /openai/v1/)
  • AZURE_AI_DEPLOYMENT = gpt-5.1-chat ✅
  • AZURE_AI_API_VERSION = 2024-12-01-preview ✅
  • AZURE_AI_MODEL_ROUTER = (empty or not set) ✅
  • AZURE_AI_KEY = (from Key Vault) ✅

Test Command

To reproduce this test:

python3 scripts/test-chat-debug.py \
  --token "YOUR_JWT_TOKEN_HERE" \
  --message "hi"
  • docs/troubleshooting/chat-backend-test-results.md - Previous test results
  • docs/troubleshooting/chat-error-diagnosis.md - Diagnostic guide
  • docs/configuration/config-alignment.md - Configuration reference
  • docs/troubleshooting/bypass-model-router.md - Disable Model Router
  • docs/troubleshooting/api-version-model-version-mismatch.md - API version issues