open/ktg-plugin-marketplace
1
0
Fork 0
Code Issues Pull requests Releases Wiki Activity Actions

fix(llm-security): template — v1 → v2 risk constants + narrative_audit block

Browse source 
Updates the HTML-comment risk-formula reference at lines 55-66 from the
stale v1 sum-and-cap formula to the v2 severity-dominated tiers that
have been authoritative in scanners/lib/severity.mjs since v7.0.0. Adds
a Narrative Audit block inside the Executive Summary section surfacing
summary.narrative_audit.suppressed_findings.{count,by_category} from
the agent's trailing JSON. The block is transparency only — it does
NOT affect risk_score, riskBand, or verdict.

Part of v7.1.1 narrative-coherence patch (plan: .claude/plans/ultraplan-2026-04-29-report-coherence.md).

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
This commit is contained in:
Kjell Tore Guttormsen 2026-04-29 12:45:28 +02:00
commit 899cb5c121
1 changed files with 36 additions and 11 deletions
Download patch file Download diff file Expand all files Collapse all files

47
plugins/llm-security/templates/unified-report.md
View file

@ -52,17 +52,29 @@ Rollback | - | - | - | - | - |
Recommendations | Y | Y | - | Y | Y | - | - | Y | -
Footer | Y | Y | Y | Y | Y | Y | Y | Y | Y
RISK SCORING (unified — all analysis types)
Formula: score = min((Critical × 25) + (High × 10) + (Medium × 4) + (Low × 1), 100)
Bands: 0-20 Low, 21-40 Medium, 41-60 High, 61-80 Critical, 81-100 Extreme
Verdict: BLOCK if Critical >= 1 OR score >= 61
WARNING if High >= 1 OR score >= 21
ALLOW otherwise
Grade: A: pass_rate >= 0.89 AND zero FAIL in cat 1,2,5 AND zero Critical
B: pass_rate >= 0.72 AND zero Critical
C: pass_rate >= 0.56
D: pass_rate >= 0.33
F: pass_rate < 0.33 OR 3+ Critical
RISK SCORING (v2 — severity-dominated, log-scaled, v7.0.0+)
See scanners/lib/severity.mjs riskScore(), verdict(), riskBand() —
this comment block is reference only; severity.mjs is authoritative.
Tiers (riskScore):
critical >= 1 → 70-95 (1=80, 2=86, 4=93, 10=95)
high only → 40-65 (1=48, 5=60, 17=65)
medium only → 15-35 (1=20, 5=28, 50=33)
low only → 1-11 (1=4, 10=11)
none → 0
Bands (riskBand): 0-14 Low, 15-39 Medium, 40-64 High, 65-84 Critical, 85-100 Extreme
Verdict: BLOCK if critical>=1 OR score>=65
WARNING if high>=1 OR score>=15
ALLOW otherwise
Grade (gradeFromPassRate, posture/audit only):
A: pass_rate >= 0.89 AND zero FAIL in cat 1,2,5 AND zero Critical
B: pass_rate >= 0.72 AND zero Critical
C: pass_rate >= 0.56
D: pass_rate >= 0.33
F: pass_rate < 0.33 OR 3+ Critical
FINDING CATEGORIES
Secrets, Injection, Permissions, Supply Chain, MCP Trust,
@ -124,6 +136,19 @@ SEVERITY CLASSIFICATION
{{EXECUTIVE_SUMMARY}}
<!-- SECTION: Narrative Audit — scan, deep-scan, plugin-audit (transparency only — does NOT adjust verdict) -->
### Narrative Audit
**Suppressed signals:** {{SUPPRESSED_FINDINGS_COUNT}} ({{SUPPRESSED_FINDINGS_BREAKDOWN}})
> Per `summary.narrative_audit.suppressed_findings`. Suppressed signals
> are raw extractor matches (entropy, frontmatter, taint) that the agent
> downgraded after context evaluation (e.g., GLSL keywords, framework
> env-var references, animation markup, SVG inline data URIs). They do
> NOT appear in the Findings sections and do NOT affect risk_score or
> verdict. The category breakdown is for reviewer transparency only.
---
<!-- SECTION: System Description — threat-model only -->