MultiTool docs

MultiTool is a monitoring and alerting platform that detects production issues as they emerge. By analyzing live service traffic with a statistical model, MultiTool automatically identifies regressions and triggers thresholdless alerts—no manual tuning required. Getting started is low friction: send your OpenTelemetry data to MultiTool and we continuously analyze your service without changing how you deploy.

The problem MultiTool solves

Alert fatigue is a common issue in mature systems. As the number of services grows, so does the amount of monitoring required. Teams typically configure alerts across metrics like error rate, CPU, memory, latency, throughput, IOPS, and disk usage, with thresholds hand-tuned to each service’s expected behavior.

When alert thresholds are too permissive, critical events can be missed and customer experience may degrade. When thresholds are too sensitive, false positives can generate excessive noise, contributing to alert fatigue.

Determining appropriate thresholds is an iterative and time-consuming process. It often involves periods of false positives and delayed detection while thresholds are refined. As systems scale and code changes are introduced more rapidly, including through AI-assisted development, maintaining accurate thresholds becomes increasingly difficult.

MultiTool takes a different approach: it's zero-config and zero-install. It learns what “normal” looks like from real production traffic and builds a model to detect meaningful changes in that behavior. Instead of static thresholds, alerts are triggered if MultiTool is confident that its seeing an anomalous dip in performance.

How MultiTool works

MultiTool ingests your service’s telemetry data and analyzes it in real time. Once data is flowing:

• Your services appear automatically in the app
• Traffic is monitored across versions and environments
• A baseline is established from healthy behavior

New traffic is continuously compared against this baseline. When MultiTool detects a significant deviation, it triggers a thresholdless alert. Each alert includes:

• A summary of what changed
• Metric visualizations for quick diagnosis
• Context across versions and environments