AI Code Quality Auditor
{{ run_id }}
← all reports
{% if banner_kind == 'dissertation' %}
✅
Dissertation result.
{{ banner_text }}
{% elif banner_kind == 'pilot' %}
⚠️
Pilot data — not yet a dissertation result.
{{ banner_text }}
{% endif %}
Five-condition comparison
Spec:
{{ provenance.spec if provenance and provenance.spec else 'unknown' }}
· {{ metrics | length }} metrics × {{ conditions | length }} conditions · click a condition to filter all charts
{% for c in conditions %}
{{ c }}
{% endfor %}
{% for s in summary %}
{{ s.metric }}
{{ '%.2f' % s.best_value }}
{{ s.unit }}
Best:
{{ s.best_condition }}
Worst: {{ s.worst_condition }} ({{ '%.2f' % s.worst_value }})
{{ s.blurb }}
{% endfor %}
Normalised comparison
Each axis is one metric, rescaled so 1.0 is the best in this report.
Leaderboard
Composite of per-metric ranks (lower is better).
{% for row in leaderboard %}
{{ row.overall_rank }}
{{ row.condition }}
avg rank {{ row.avg_rank }}
{% if row.wins > 0 %}🏆 {{ row.wins }}{% endif %}
{% endfor %}
Heatmap
Green = best for that metric, red = worst. Normalised per row.
Per-metric drill-down
Click a tab to inspect one metric.
{% for m in metrics %}
{{ m }}
{% endfor %}
{% for m in metrics %}
{{ m }}
{{ blurbs[m] }}
Unit:
{{ units[m] or '—' }} ·
Lower is better
Condition
Value
Rank
{% for c in conditions %}
{{ c }}
{{ '%.3f' % pivot[m][c] }}
{{ ranks[m][c] }}
{% endfor %}
{% endfor %}
{% if provenance %}
Provenance metadata
{{ provenance | tojson(indent=2) }}
{% endif %}