Stack traces are useful, but they are not always easy to act on quickly.
When something breaks, you usually want more than the exception name. You want to know the likely root cause, how serious it is, where to look, and what fix to try first.
This Python example turns a stack trace into structured debugging JSON using Telnyx AI Inference.
The Flask app exposes:
POST /explain
GET /analyses
GET /analyses/<id>
GET /health
POST /explain
accepts a stack trace, plus optional language and runtime context:
{
"language": "python",
"context": "Flask production server with gunicorn",
"stack_trace": "Traceback (most recent call last):\n File \"app.py\", line 42..."
}
The app sends the trace to Telnyx AI Inference and asks for structured JSON:
{
"root_cause": "The outbound HTTP call is failing because the downstream service is unreachable.",
"severity": "high",
"confidence": 0.91,
"likely_culprit": "app.py:42",
"suggested_fix": "Add timeout handling and retry logic around the request.",
"fix_snippet": "resp = requests.post(url, timeout=15)",
"related_errors": ["requests.exceptions.Timeout"],
"prevention": "Set explicit timeouts for outbound HTTP calls."
}
Plain text explanations are nice for humans. Structured output is useful for apps.
With this response shape, you could:
The example stores recent analyses in memory and lets you retrieve them by ID.
Clone the examples repo:
git clone https://github.com/team-telnyx/telnyx-code-examples.git
cd telnyx-code-examples
git switch feat/error-explainer-python
cd error-explainer-python
Create your .env
file:
cp .env.example .env
Add your Telnyx API key:
TELNYX_API_KEY=your_telnyx_api_key
AI_MODEL=moonshotai/Kimi-K2.6
HOST=127.0.0.1
Install and start:
pip install -r requirements.txt
python app.py
Try the health endpoint:
curl http://localhost:5000/health
Explain an error:
curl -X POST http://localhost:5000/explain \
-H "Content-Type: application/json" \
-d '{
"language": "python",
"context": "Flask production server",
"stack_trace": "KeyError: user_id"
}' | python3 -m json.tool
This is a small example, but it maps pretty cleanly to real developer workflows:
The useful part is not only that a model can explain an error. It is that the app gets back a predictable object it can route, store, display, or review.
Telnyx AI skills and toolkits: https://github.com/team-telnyx/ai
Telnyx AI Inference docs: https://developers.telnyx.com/docs/inference
Chat Completions API: https://developers.telnyx.com/api/inference/chat-completions
Telnyx Portal: https://portal.telnyx.com/