• LOGS
  • TRACES
  • METADATA
  • HUMAN

How to send OTLP telemetry to Highlight with your AI

Connect Highlight to your AI client once. After that, your agent handles the heavy lifting of processing telemetry data.

Ask AI about this page

Short answer

How can I send OTLP telemetry to Highlight using my AI?

Your AI uses the Highlight MCP to ingest structured OTLP logs and traces directly. This allows your agent to visualize request flows and maintain rich metadata without manual parsing. You'll get a clear view of how data moves through your system.

Where every issue lands.

Telemetry outcomes

The AI processes incoming data to give you these specific views.

LOGS

Structured log ingestion

The AI pulls in OTLP JSON logs. You get organized, searchable text that keeps all your custom attributes.

TRACES

Visual request flows

Traces are ingested to show how requests move. Your agent maps the path from start to finish.

METADATA

Context preservation

The AI keeps your metadata attached to every event. You won't lose the specific tags that make debugging useful.

HUMAN

Manual triage

The AI identifies the patterns, but you decide which bugs to fix. It flags the anomalies for your review.

The workflow

What your AI does when telemetry arrives.

The AI acts as the bridge between your telemetry stream and your debugging session.

  1. Log ingestion

    The AI pulls in structured JSON logs from your stream. It ensures every attribute stays mapped correctly.

    ingest_otlp_logs

  2. Trace ingestion

    The agent ingests traces to build a map of your service calls. This links individual spans together.

    ingest_otlp_traces

  3. Flow mapping

    The AI looks at the ingested traces to reconstruct the request path. It identifies where a sequence breaks.

    ingest_otlp_traces

  4. Contextual analysis

    The agent combines logs and traces to find the root cause. It uses the metadata to provide specific details.

    ingest_otlp_logsingest_otlp_traces

Try it

Copy these to start.

Use these prompts to kick off your analysis.

Starting points

These are basic ways to start a conversation. Swap out the specific service or trace ID with your own data.

Highlight Connector

You're all set. Choose your MCP client and follow the setup instructions.

Connector linkhttps://edge.vinkius.com/vk_preview_NSOfm3XYtPzM2Hc98HN4tUV10V3ARA5UYzOvTYCR/mcp

Claude Desktop

Follow the steps below to connect in seconds.

  1. 1In Claude Desktop, open Settings → Connectors.
  2. 2Click “Add custom connector” and paste the connector link above as the remote MCP server URL.
  3. 3Click Add and start a new chat — Highlight capabilities are ready to use.
Configuration · claude_desktop_config.jsonCopy
{
  "mcpServers": {
    "highlight-session-replay-ux-mcp": {
      "url": "https://edge.vinkius.com/vk_preview_NSOfm3XYtPzM2Hc98HN4tUV10V3ARA5UYzOvTYCR/mcp"
    }
  }
}
Copy into chat04

  • Show me the request flow for the last failed trace in the production service.

  • Analyze the OTLP logs from the auth-service to find why latency spiked.

  • Map the trace path for user ID 8829 to see where the bottleneck is.

  • Summarize the error logs from the payment-gateway service from the last hour.

  • Claude
  • ChatGPT
  • Cursor
  • VS Code
  • Windsurf
  • Claude Code
  • JetBrains
  • Cline

Start here

Connect Highlight once, then ask.

Just click the link to connect. Your credentials stay encrypted, and you can start chatting with your telemetry immediately.

Connect Highlight to your AI

FAQ

How the Highlight connector behaves

  • 01

    Can the AI change my Highlight configuration?

    No. The AI can only read and ingest telemetry data. It cannot modify your Highlight settings or account setup.

  • 02

    Does the AI delete any of my logs?

    No. The tools only ingest and read data. The AI has no capability to delete or modify existing telemetry in Highlight.

  • 03

    Can I use this with any OTLP source?

    Yes, as long as the data follows the OTLP standard and is accessible via the ingest tools.

  • 04

    Will the AI handle my trace metadata?

    Yes. The ingestion process is designed to keep your structured JSON metadata intact for the AI to use.

  • 05

    Do I need to manually format my logs for the AI?

    No. If your logs are already in structured OTLP JSON format, the AI can ingest them directly.

  • More questions about Highlight? See everything the Highlight Connector can do

Connect Highlight to Claude, Cursor, ChatGPT & more