## Why use Opik for observability

Debugging LLM applications without observability means guessing. You see the final output but not why the model hallucinated, which retrieval step returned irrelevant context, or where latency spiked.

With Opik, you can:

- **See the full execution path** of every request — from user input through tool calls and LLM completions to the final response
- **Root-cause production issues fast** — filter and search traces by status, latency, cost, or custom tags to find the problem in seconds
- **Track costs and latency over time** — monitor token usage and spending across models and providers
- **Capture multi-turn conversations** — group related traces into threads to understand how interactions evolve across turns
- **Close the feedback loop** — attach human or automated scores to traces and use them to drive evaluations

## What you can capture

[Traces & spans\
\
Full execution trees with inputs, outputs, timing, and metadata for every step](/content/docs/opik/tracing/concepts/index.html) [Conversations\
\
Multi-turn threads that group related traces into coherent sessions](/content/docs/opik/tracing/advanced/log_chat_conversations/index.html) [Cost tracking\
\
Token usage and spending broken down by model, provider, and trace](/content/docs/opik/tracing/advanced/cost_tracking/index.html) [Media & attachments\
\
Images, audio, video, and files logged alongside your traces](/content/docs/opik/tracing/advanced/log_multimodal_traces/index.html) [User feedback\
\
Qualitative and quantitative scores attached to individual traces](/content/docs/opik/tracing/advanced/annotate_traces/index.html) [Agent graphs\
\
Visual execution graphs showing how your agent’s steps connect](/content/docs/opik/tracing/advanced/log_agent_graphs/index.html)

## How it works

### Connect your project

Run `opik connect` from your agent’s directory to pair it with Opik:

```
opik connect --project <YOUR_PROJECT_NAME>
```

### Instrument your code

The fastest way to add tracing is with [opik-skills](https://github.com/comet-ml/opik-skills) — install the skill and let your coding agent handle the rest:

```
npx skills add comet-ml/opik-skills
```

Then ask your coding agent:

```
Instrument my agent with Opik using the /instrument command.
```

This works with Claude Code, Cursor, Codex, OpenCode, and other coding agents. You can also instrument manually with the SDK:

PythonTypeScript

```
import opik

@opik.track
def my_agent(user_message):
    context = retrieve_context(user_message)
    response = call_llm(user_message, context)
    return response
```

### View traces in the dashboard

Every request creates a trace with detailed span-level information. You can inspect the full execution tree, see inputs and outputs at each step, and filter by duration, cost, status, or tags.

### Analyze and improve

Use traces to debug failures, identify slow steps, and track quality over time. Attach feedback scores, run evaluations against datasets, and use [Ollie](/content/docs/opik/tracing/debug-agents/index.html) — Opik’s AI assistant — to help root-cause issues automatically.

## Integrations

Opik has first-class support for 30+ frameworks in Python, TypeScript, and OpenTelemetry — so you can start capturing traces without changing how your application is built.

## Next steps

- [Getting started](/content/docs/opik/tracing/getting-started/index.html) — Add observability to your agent in minutes
- [Concepts](/content/docs/opik/tracing/concepts/index.html) — Understand traces, spans, threads, and feedback scores
- [Debugging agents with Ollie](/content/docs/opik/tracing/debug-agents/index.html) — Use AI-assisted root-cause analysis
- [Cost tracking](/content/docs/opik/tracing/advanced/cost_tracking/index.html) — Monitor token usage and spending