Flamehaven commited on
Commit
d20aef9
·
verified ·
1 Parent(s): ecf847a

release: v1.8.2 ICH M15 citation alignment

Browse files
Files changed (1) hide show
  1. docs/UI_HTML_REPORT.md +20 -3
docs/UI_HTML_REPORT.md CHANGED
@@ -26,14 +26,31 @@ Example:
26
 
27
  | Section | ID | Content |
28
  |---------|-----|---------|
29
- | **Executive Summary** | `#s1` | Score gauge, 5-metric stat grid, notable risks, T0 alert banner |
30
- | **Decision Path** | `#s2` | Stage bars (S1/S2R/S3/S4), score formula, and policy/configuration explainer |
31
  | **Code Integrity & Contract** | `#s3` | C1–C6 + CC1–CC3 expandable cards, with C4/C5/C6 split visible |
32
- | **AIRI Risk Coverage** | `#s4` | MIT AI Risk Repository donut + covered/gaps toggle table + domain filter cards |
33
  | **Evidence Detail** | `#s5` | Full evidence ledger with severity filter chips |
34
 
35
  Navigation between sections via a **sticky top navbar** with scroll-spy active state.
36
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
37
  ## Interactive Features
38
 
39
  ### Score Gauge
 
26
 
27
  | Section | ID | Content |
28
  |---------|-----|---------|
29
+ | **Executive Summary** | `#s1` | Score gauge, 5-metric stat grid, notable risks, Tier Lock alert banner, Classification Applied state |
30
+ | **Decision Path** | `#s2` | Stage bars (S1/S2R/S3/S4 with Stage 3 raw formula), score formula, and policy/configuration explainer |
31
  | **Code Integrity & Contract** | `#s3` | C1–C6 + CC1–CC3 expandable cards, with C4/C5/C6 split visible |
32
+ | **AIRI Risk Triggers** | `#s4` | MIT AI Risk Repository donut + covered/gaps toggle table + domain filter cards |
33
  | **Evidence Detail** | `#s5` | Full evidence ledger with severity filter chips |
34
 
35
  Navigation between sections via a **sticky top navbar** with scroll-spy active state.
36
 
37
+ ### Tier Lock Alert (Section 1)
38
+
39
+ When a score ceiling is active, a colored alert banner appears at the top of the Executive Summary:
40
+
41
+ - **`Tier Lock [T0-FLOOR]`** (red): CA-DIRECT classification with no disclaimer. Score ceiling at 39. Must be resolved before any tier advancement.
42
+ - **`Tier Lock [CA-CAP]`** (amber): Clinical-adjacent surface without non-clinical boundary. Score ceiling at 69 (T2 maximum). Adding a non-diagnostic disclaimer resolves this lock.
43
+
44
+ ### Classification Applied (Section 1)
45
+
46
+ The Policy Boundary card in Section 1 now shows the applied classification state:
47
+
48
+ ```
49
+ Classification applied: ca_severity=CA-INDIRECT | score_cap=69 | t0_floor=clear
50
+ ```
51
+
52
+ This makes the active cap and severity class inspectable without reading the raw JSON.
53
+
54
  ## Interactive Features
55
 
56
  ### Score Gauge