# Highlight MCP for AI Agents AI Agent Connect > Highlight (Session Replay & UX) MCP lets your AI agent push raw logs, OTLP logs, and OTLP traces directly into your Highlight project. It connects your technical observability data to your session replays so you can see exactly what happened in the backend when a user hit a snag. ## Overview - **Category:** developer-tools - **Price:** Free - **Endpoint:** https://edge.vinkius.com/vk_preview_NSOfm3XYtPzM2Hc98HN4tUV10V3ARA5UYzOvTYCR/ai-agent-connect - **Tags:** session-replay, ux-monitoring, observability, otlp, logging ## Description Stop jumping between your terminal, your log aggregator, and your session replay tool to figure out why a user just saw a blank screen. This Connector lets you feed all that technical data into Highlight using just your AI client. Instead of manual uploads or complex configuration, you can tell your agent to grab a specific set of logs or a trace and push it right into your dashboard. It's a way to keep your observability data organized without the friction of manual data entry. If you're looking for a way to make your monitoring more actionable, this is a solid addition to the Vinkius catalog. You get to see the connection between a specific backend error and the actual user experience in one place. It's about getting the full picture of a bug without the headache of hunting down the right IDs across multiple tabs. ## Tools ### ingest_raw_log Send a simple text log message to your Highlight dashboard. This helps when you need to quickly record a specific event or error string. ### ingest_otlp_logs Push structured OTLP JSON logs into Highlight. Use this to keep all your metadata and context intact for complex systems. ### ingest_otlp_traces Send OTLP JSON traces to Highlight. This lets you visualize the entire request flow and pinpoint where the lag is happening. ## Prompt Examples **Prompt:** ``` I just got a 'Database Timeout' error in my local logs. Can you send that to my Highlight project? ``` **Response:** ``` I've sent the 'Database Timeout' error to your Highlight project. You can now see it in your dashboard alongside any related user sessions. ``` **Prompt:** ``` Send the last 5 OTLP logs from our auth service to Highlight. ``` **Response:** ``` I'm processing the OTLP logs for the auth service. They've been successfully pushed to your Highlight dashboard. ``` **Prompt:** ``` The checkout page is slow. Send the latest traces to Highlight so we can check the request flow. ``` **Response:** ``` I've sent the latest OTLP traces to Highlight. You can now analyze the request spans and performance metrics in your project. ``` ## Capabilities ### Push raw text logs to Highlight Send simple log messages from your backend or terminal directly to your dashboard. ### Upload structured OTLP JSON logs Keep your metadata intact by pushing structured logs into your project. ### Send OTLP traces to your dashboard Visualize request flows and find performance bottlenecks by pushing traces. ### Connect backend errors to user sessions Link technical failures to actual user interactions in one view. ### Automate observability data ingestion Move data from your dev environment to Highlight without manual steps. ## Use Cases ### Debugging a production 500 error A developer sees a '500' error in a local terminal and wants to see if it happened in production. They ask the agent to send the raw log to Highlight. ### Verifying a new deployment A DevOps engineer needs to verify if a new deployment is causing latency. They have the agent push OTLP traces to the dashboard. ### Finding checkout friction A product manager wants to know why users are dropping off at checkout. They ask the agent to pull OTLP logs for that specific page. ### Tracking down an edge case An engineer finds a weird edge case in a microservice. They use the agent to ingest the raw log to see it in the session replay. ## Benefits - See the direct link between a backend error and a user's session by using ingest_otlp_logs to keep metadata together. - Skip the manual copy-pasting of error messages by using ingest_raw_log to push text logs instantly. - Visualize slow requests across your stack using ingest_otlp_traces to find bottlenecks faster. - Keep your dashboard updated with the latest OTLP data without manual exports or complex scripts. - Give your product team a clear view of UX friction by linking logs to actual user interactions. ## How It Works The bottom line is you get a direct line from your logs to your user sessions without the manual middleman. 1. Subscribe to the Connector and provide your Highlight Project ID. 2. Tell your AI agent which logs or traces to send from your environment. 3. View the correlated data in your Highlight dashboard. ## Frequently Asked Questions **How does Highlight MCP help with session replays?** It connects your backend logs to your replays so you can see the exact error a user encountered in real time. **Can I send logs from my terminal to Highlight using this?** Yes, you can use the raw log tool to push text directly from your environment to your dashboard. **What is the benefit of using OTLP logs with Highlight?** It keeps your logs structured, meaning you don't lose important metadata when looking at a session. **Does Highlight MCP support traces?** Yes, it allows you to send OTLP traces to see the full request flow and find bottlenecks. **Is this good for product managers?** It's great for correlating backend issues with user behavior to find UX friction points. **How do I get my logs into Highlight?** You just need to provide your Project ID and tell your agent which logs or traces to send. **How can I send a basic text log message to my dashboard?** You can use the `ingest_raw_log` tool. Simply provide the service name and the message content, and it will be sent directly to Highlight. **Does this server support structured OpenTelemetry logs?** Yes! Use the `ingest_otlp_logs` tool to send structured logs in OTLP JSON format. Ensure your payload includes the project ID attribute. **Can I visualize request traces using this integration?** Absolutely. The `ingest_otlp_traces` tool allows you to send OTLP JSON traces to Highlight, helping you track request spans and performance.