# bryel

> LLM & agent observability — OpenTelemetry-native tracing for any framework.

## Docs

- [Concepts](https://docs.bryel.ai/concepts.md): How bryel models agent runs as traces, interactions, and sessions.
- [Coding agents](https://docs.bryel.ai/guides/coding-agents.md): Connect Claude Code, Cursor, or any MCP client to bryel to query traces, inspect dataset quality, and build fine-tune datasets from your editor.
- [Datasets](https://docs.bryel.ai/guides/datasets.md): Curate production traces into supervised fine-tuning (SFT) datasets, judge their quality, and export them for training.
- [Feedback](https://docs.bryel.ai/guides/feedback.md): Record 👍/👎, scores, comments, and corrections against a generation with @bryel/feedback.
- [bryel](https://docs.bryel.ai/introduction.md): OpenTelemetry-native observability for LLM apps and agents — ship traces from any framework to bryel.
- [Quickstart](https://docs.bryel.ai/quickstart.md): Instrument an app, send a trace, and see it in bryel in under five minutes.
- [Browser (JavaScript)](https://docs.bryel.ai/sdk/browser.md): Trace an AI agent that runs in the browser, with a write-only key locked to your domains.
- [HTTP API](https://docs.bryel.ai/sdk/http.md): Send traces to bryel from any language over OTLP/HTTP — no bryel SDK required.
- [Integrations](https://docs.bryel.ai/sdk/integrations.md): Every provider SDK, agent SDK, and framework bryel captures — and the one line that turns each on.
- [Python](https://docs.bryel.ai/sdk/python.md): Trace a Python LLM app or agent — one call, then any OpenInference instrumentor.
- [TypeScript](https://docs.bryel.ai/sdk/typescript.md): Trace a Node/TypeScript LLM app — the Vercel AI SDK, any OpenTelemetry app, or a provider SDK directly.