dashboard

Run the attune-rag dashboard against the default corpus with one command:

attune-rag dashboard

This runs the full three-stage pipeline — snapshot → render → display — and exits with code 0 on success.

How it works

The dashboard pipeline has three stages:

  1. Refresh — build_snapshot() benchmarks your corpus and returns a snapshot dict. If queries.yaml is missing, it returns a partial snapshot with an error field instead of raising.
  2. Render — render(out, snapshot) writes a self-contained HTML file to out with the snapshot embedded as JSON.
  3. Display — display(snapshot) pretty-prints the snapshot to the terminal using Rich.

Run the pipeline in Python

from pathlib import Path
from attune_rag.dashboard.refresh import build_snapshot
from attune_rag.dashboard.render import render
from attune_rag.dashboard.show import display

snapshot = build_snapshot(corpus_package="attune_help")
render(out=Path("dashboard.html"), snapshot=snapshot)
display(snapshot)

You should see Rich-formatted output in your terminal and a dashboard.html file written to the current directory.

Use a custom corpus or queries file

from pathlib import Path
from attune_rag.dashboard.refresh import build_snapshot

snapshot = build_snapshot(
    corpus_package="my_corpus",
    queries_path=Path("path/to/queries.yaml"),
)

If queries_path does not exist, build_snapshot() returns a partial snapshot — check for an "error" key in the result before passing it downstream.

Expected output

A successful display() call prints a Rich table to the terminal. A successful render() call produces an HTML file whose source contains the embedded snapshot where the placeholder __ATTUNE_SNAPSHOT__ was replaced with the snapshot JSON.

dashboard.html  ← self-contained HTML report
<Rich table>    ← terminal summary

Next: Open dashboard.html in a browser to review the full rendered report.