tracewayapp/traceway — reverse-engineered prompt

Build me a self hosted observability app called Traceway that feels like one place to understand what is happening in production and how to fix it fast.

I want a dashboard that can ingest OpenTelemetry data directly, without needing a bunch of extra setup, then show logs, traces, metrics, exceptions, session replay, and AI request tracing together around the same trace or issue. Make it easy to search logs, inspect span waterfalls, see endpoint latency and error rate, group exceptions into issues, and view host or app metrics in charts. Include alerting to common channels, multi tenant organizations with roles, and a simple way to monitor background jobs too.

It should be easy to run locally with Docker Compose, and there should also be an embedded mode for Go apps using SQLite. Please include source map and mobile crash symbolication so production errors resolve back to real code. If the repo already has a CLI and agent friendly skills, wire those up so an assistant can set it up and query telemetry safely. Look up current docs online if you need to.