# bryel

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

## Docs

- [@bryel/feedback](https://docs.bryel.ai/api-reference/feedback.md): Record user feedback against a generation or session.
- [@bryel/sdk-core](https://docs.bryel.ai/api-reference/sdk-core.md): Framework-agnostic core: OTLP exporter + span processors.
- [@bryel/semconv](https://docs.bryel.ai/api-reference/semconv.md): OpenInference conventions + bryel extensions.
- [@bryel/vercel](https://docs.bryel.ai/api-reference/vercel.md): Vercel AI SDK adapter.
- [Any OpenTelemetry app](https://docs.bryel.ai/frameworks/opentelemetry.md): Ship to bryel from any OpenTelemetry-instrumented app.
- [OpenAI · Anthropic · LangChain](https://docs.bryel.ai/frameworks/other.md): Trace direct provider SDKs by shipping their OpenTelemetry spans to bryel.
- [Vercel AI SDK](https://docs.bryel.ai/frameworks/vercel.md): Trace the Vercel AI SDK with one processor and a per-call flag.
- [Query from your coding agent](https://docs.bryel.ai/guides/coding-agents.md): Connect bryel to Claude Code, Cursor, or any MCP client and query your traces with BQL.
- [Concepts](https://docs.bryel.ai/guides/concepts.md): How bryel models agent runs: traces, interactions, and sessions.
- [Feedback](https://docs.bryel.ai/guides/feedback.md): Record 👍/👎, scores, comments, and corrections against a generation.
- [Installation](https://docs.bryel.ai/installation.md): Install the bryel SDK packages for your stack.
- [Introduction](https://docs.bryel.ai/introduction.md): OpenTelemetry-native tracing for LLM apps and agents — ship to bryel from any framework.
- [Quickstart](https://docs.bryel.ai/quickstart.md): Get your first trace into bryel in a couple of minutes.