Overview
What the Telflo AI assistant can do and how to use it.
The AI assistant is a chat panel built into the editor. It can design pipelines from a description, generate and modify your configuration directly on the canvas, and answer OpenTelemetry questions — all without leaving Telflo.
Opening the assistant
Open the AI tab in the editor. The tab is only present when AI is enabled for you:
- AI must be turned on in Settings → AI (the Enable AI Features toggle).
- Depending on your account, AI access may need to be requested first.
See AI access for how to enable and request access.
What you can ask
Create configurations
- "Create a pipeline that receives OTLP and exports to Jaeger."
- "Build a Kubernetes monitoring setup with metrics and logs."
Modify the current configuration
- "Add a batch processor to the current pipeline."
- "Change the OTLP exporter endpoint to
otel.example.com:4317."
Ask questions
- "What's the difference between OTLP and OTLP/HTTP?"
- "How do I configure tail sampling?"
How it works with your canvas
The assistant can read and edit the configuration you're working on. It uses two tools behind the scenes:
| Tool | What it does |
|---|---|
| Read current YAML | Reads your current configuration so its answers and edits are grounded in what you actually have. |
| Update YAML | Applies changes to your configuration — the canvas and YAML update to reflect them. |
When the assistant updates your configuration, the change appears on the canvas just as if you'd made it yourself, so you can review, tweak, undo, and save it normally.
Modes
You steer the assistant by choosing a mode — Plan, Build, or Ask — in the chat composer. Each mode is tuned for a different task. See Modes.
Conversations
Your chats are saved, and Telflo generates a title for each conversation so you can find it again later.
Safety
Every request and response passes through guardrails that block unsafe or off-topic content and protect against leaking sensitive data.