Free tools Windows power users keep installed
One-click scans. No signup required.
JetBrains announced Tracy on March 11, 2026: an open-source Kotlin library for tracing AI application activity, including model calls, tool executions, and custom code. It uses OpenTelemetry, supports Kotlin and Java projects, and can export telemetry to destinations such as Langfuse and W&B Weave. By default, Tracy redacts or omits prompt and response content; capturing it requires an explicit opt-in.
What JetBrains Tracy does
Tracy adds observability to AI-powered applications by recording activity as OpenTelemetry spans. The goal is to show more than just model requests: developers can trace an agent invocation, its language-model calls, the tools it runs, and other application logic. JetBrains describes Tracy as an open-source Kotlin library that adds production-grade observability to AI applications in minutes.
It follows OpenTelemetry’s Generative AI semantic conventions and offers several ways to create traces: instrumentation for supported AI and HTTP clients, manual spans around arbitrary code, and Kotlin annotation-based tracing with @Trace. In JetBrains’ example, withSpan creates a span for an agent operation while instrument(client) instruments an LLM client.
Tracing tool execution matters because an agent’s behavior is distributed across multiple operations. Tracy can trace an annotated interface method so implementing tool classes inherit tracing behavior, avoiding repeated instrumentation code.
What’s actually slowing this PC down?
Pick the symptom - the matching free tool is one click away.
#1 Best Overall
How Tracy differs for Kotlin and Java
| Language | Tracing approach | What it provides |
|---|---|---|
| Kotlin | @Trace annotations with Tracy’s compiler plugin, or manual APIs such as withSpan |
Annotation-based tracing can record execution timing, inputs, and outputs; manual spans allow explicit boundaries and metadata. |
| Java | Manual tracing APIs such as withSpan |
Developers define span boundaries and metadata explicitly. Annotation-based tracing is not supported for Java. |
Although Tracy is a Kotlin library, JetBrains documents its shared API and tracing modules for Kotlin or Java projects. The key language difference is the compiler-plugin feature: Java developers use manual spans rather than @Trace.
Requirements and supported clients
JetBrains’ repository, checked September 30, 2026, lists these requirements and supported ranges:
Rank #2
| Component | Documented support |
|---|---|
| Kotlin | 2.0.0 or newer |
| Java | 17 or newer |
| OpenTelemetry | 1.2 or newer, when OpenTelemetry is already installed |
| OpenAI SDK | 1.x–4.x |
| Anthropic SDK | 1.x–2.x |
| Gemini SDK | 1.8.x–1.38.x; earlier versions are unsupported |
These are implementation details that can change. Check the Tracy repository for the current release and compatibility information before adding it to a project. The first public release listed there is 0.1.0.
Adding Tracy to a project
The repository documents Gradle setup using the org.jetbrains.ai.tracy plugin and the modules required for the integrations you use. Available modules include tracy-core, tracy-openai, tracy-anthropic, tracy-gemini, and tracy-ktor. Maven coordinates are also documented. Consult the repository’s setup instructions for the exact configuration for your build tool and versions.
Do these 3 things before closing this tab:
1Repair Windows errors before they cause bigger problems2Scan for outdated or missing drivers - takes under a minute3Clear out junk files and repair common Windows errorsAfter setup, instrument supported clients and add manual spans around the agent and tool operations you want to inspect. For Kotlin, the compiler plugin also enables annotation-based tracing. For Java, use manual spans to define those boundaries.
Prompt and response capture is opt-in
Tracy’s default is to record metadata while redacting or not capturing sensitive user and assistant content. The README describes the default placeholder as REDACTED. Inputs and outputs can be enabled independently, either by calling TracingManager.traceSensitiveContent() or by setting the corresponding environment variables: TRACY_CAPTURE_INPUT=true for inputs and TRACY_CAPTURE_OUTPUT=true for outputs.
Enabling content capture changes what telemetry may contain. Review the data your application handles and your observability backend’s access and retention controls before opting in, especially if prompts or responses can include personal, confidential, or otherwise sensitive information.
Exporting traces to Langfuse, Weave, and other backends
Tracy can send telemetry to OpenTelemetry-compatible backends. JetBrains names Jaeger, Zipkin, and Grafana as examples, and the repository documents exporter configuration for Langfuse, W&B Weave, console output, and files. The specific configuration depends on the chosen destination; use the repository’s exporter examples rather than assuming every backend uses the same settings.
Recommended Free Tools
Best Value
When Tracy may fit an application
Tracy is worth evaluating when an application needs traces across model requests, tool calls, and internal agent logic, rather than visibility into client calls alone. Its OpenTelemetry foundation also gives teams a choice of compatible telemetry destinations.
- Instrumentation scope: Decide whether you need model-client traces only or visibility into internal agent and tool functions too.
- Language workflow: Kotlin projects can use annotations and the compiler plugin; Java projects need manual spans.
- Data handling: The default avoids capturing prompt and response content, while opt-in capture can expose sensitive information in telemetry.
- Compatibility: Confirm your Kotlin or Java version, AI SDK versions, and any existing OpenTelemetry installation against the repository’s current requirements.
- Export destination: Check whether the documented exporter setup matches your chosen backend and deployment.
JetBrains’ announcement and repository do not publish adoption, performance, or market-size statistics, so those are not established measures for comparing Tracy with other tracing approaches.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
