{# Live Dashboard - Coverage — pixel-match port of the Claude Design bundle ``guarding-oss/project/Live Dashboard - Coverage.html``. This template is intentionally a self-contained document (it does NOT extend base.html) because the design ships its own topbar/footer chrome with the URL chip, locality pill, and bespoke score-banner aesthetic. Matching the design precisely means owning the here. Five Jegan-comment corrections from the design bundle are already baked into the markup: * top tabs removed (the

in the design is hidden=true) * "no telemetry" wording removed from the locality pill * findings list shows ALL findings paginated (15 per page default) * default sort is criticality descending * coverage cells link to ``#findings`` so clicking jumps to the table #} {{ page_title|default("Coverage") }} · AgentGuardian {% if is_running %}{% endif %}

Scan {{ scan_id }} {% if is_running %} Live · {{ elapsed_label }} elapsed {% else %} Done · {{ elapsed_label }} elapsed {% endif %} {% if scan %} Target {{ scan.target_ref }} Adapter {{ scan.target_mode }} Tier {{ scan.tier.value }} {% endif %}

Overview Transcripts Export

Aivss tier-weighted{% if is_running %}, provisional{% endif %}

Critical Poor Warning Good Excellent {{ aivss }}

040608090100

{% if findings_total == 0 %} No findings yet. The target is defending every attempt so far. {% elif tile_sev.critical > 0 %} {{ tile_sev.critical }} unresolved critical finding{{ "s" if tile_sev.critical != 1 }} are pulling the aggregate down. {% else %} {{ findings_total }} finding{{ "s" if findings_total != 1 }} recorded across {{ tile_unique_probes }} unique probe{{ "s" if tile_unique_probes != 1 }} ({{ tile_attempts_fired }} attempt{{ "s" if tile_attempts_fired != 1 }} total). {% endif %} {% if is_running %}The swarm is still running — this score is provisional.{% endif %}

{{ findings_total }} findings

{{ "{:,}".format(tile_attempts_fired) }} attempts fired across {{ rows[0].cells|length }} ASI agents

{{ "{:,}".format(tile_unique_probes) }} unique probes distinct seeds attempted

{% if tile_tokens_total >= 1000 %}{{ "{:.0f}k".format(tile_tokens_total / 1000) }}{% else %}{{ tile_tokens_total }}{% endif %} tokens spent attacker + judge

$ {{ "{:.2f}".format(tile_cost) }} cost USD, model-priced

{{ tile_agents_active }} agents active recon + 10 specialists

{{ tile_remaining }} {% if is_running %}remaining{% else %}duration{% endif %} {{ elapsed_label }} elapsed

Coverage by framework

Click any cell to drill into its findings · 5 frameworks shown

In scope All 5 Critical Failing

Scale excellent · ≥ 90 good · 80–89 attention · 70–79 failing · < 70 in progress

{% for row in rows %}

{% for c in row.cells %} {{ c.name }} {% if c.sub %}{{ c.sub }}{% endif %} {% if c.score_pct is not none %}{{ c.score_pct }}%{% else %}—{% endif %} {{ c.count_label }} {% if c.has_warn %}{% endif %} {% endfor %}

{% endfor %}

Findings

All findings so far.

{{ findings_total }} finding{{ "s" if findings_total != 1 }} · sorted by criticality by default · filter to drill

Export SARIF Export JSON

{% if findings_rows %} {% for f in findings_rows %} {% endfor %}

Severity	When	ASI	Finding	Frameworks	Remediation
{{ f.severity_label }}	{{ f.when_time }}{{ f.when_agent }}	{{ f.asi }}	{{ f.summary }}	{% for tag, is_probe in f.tags %} {{ tag }} {% endfor %}	`{{ f.remediation_id }}` ↗ {{ f.remediation_hint }}

{% else %}

No findings match the current filter.

{% endif %} {% if filtered_total > 0 %} {% set first_idx = (page - 1) * per_page + 1 %} {% set last_idx = first_idx + (findings_rows|length) - 1 %} {% endif %}