← Agent Platform · Prompt Deck · Pictures

What do the 8 dimensions actually measure — and what does a healthy reading look like?

The 8 Dimensions​

The VVFL Dashboard instrument reads enforcement health across 8 dimensions. Each dimension is an auditor — a function that takes evidence and returns findings.

Source: A&ID Instrument Registry — VVFL Dashboard row (line 151).

Dimension Detail​

1. Generator​

2. Template​

3. Rule​

4. Skill​

5. Agent​

6. Platform​

7. Virtue​

8. Pattern​

Audit Output Schema​

Every auditor produces findings in this shape. Forward-compatible with Phase 3 receipts.

{
  "dimension": "generator|template|rule|skill|agent|platform|virtue|pattern",
  "severity": "info|warning|critical",
  "finding": "Human-readable description",
  "evidence": {
    "file": "path/to/file",
    "line": 42,
    "expected": "what should be there",
    "actual": "what is there",
  },
  "routing": {
    "action": "fix|escalate|create",
    "target": "path/to/artifact",
    "owner": "role or team",
  },
  "gap_type": "gate-bypass|template-bloat|sequence-violation|interface-drift|demand-absence",
}

Gaps to Dimensions​

The five engineering gaps map to dimensions that catch them.

Story Contract​

Stories are test contracts. Tests must be RED before implementation starts. GREEN = value delivered.

S1 — Platform auditor reads real hook failures​

Trigger: drmg audit --dry-run runs against a repo where .claude/hooks/failure-log.jsonl contains ≥1 failure entry

Checklist:

• Output JSON has findings[] with ≥1 entry where dimension: "platform"
• evidence.source references failure-log.jsonl (not a hardcoded stub)
• severity is warning or critical — not info
• Test seeds a real failure-log.jsonl and reads it — hardcoded stub does not pass

Forbidden: Stub returning hardcoded findings passes. Auditor produces findings without reading failure-log.jsonl.

Evidence: integration — drmg/__tests__/story-s1-platform.spec.ts

Commission Result: ⬜ PASS / ⬜ FAIL Notes: (findings)

S2 — Routing fields resolve to real artifacts​

Trigger: An audit finding is produced with a routing field

Checklist:

• finding.routing.target resolves to an existing file path on disk
• finding.routing.owner matches a value in the AssignedTeam enum
• Finding with routing.target: "" or routing.owner: "unknown" is rejected as invalid

Forbidden: Finding with empty target or "unknown" owner accepted as valid.

Evidence: unit — drmg/__tests__/story-s2-routing.spec.ts

Commission Result: ⬜ PASS / ⬜ FAIL Notes: (findings)

S3 — Template auditor flags null prdRef as critical​

Trigger: A plan record exists in DB with prdRef: null

Checklist:

• auditTemplate([plan]) returns ≥1 finding where severity: "critical"
• evidence.field: "prdRef" named in the finding
• evidence.actual: null present
• Plan with null prdRef does not produce zero findings or info-only findings

Forbidden: Plan with null prdRef produces no finding, or only info severity.

Evidence: integration — drmg/__tests__/story-s3-template.spec.ts

Commission Result: ⬜ PASS / ⬜ FAIL Notes: (findings)

S4 — Generator auditor detects enum drift​

Trigger: pgEnum in schema has 3 values; TypeScript union in generated types has 5 values

Checklist:

• auditGenerator(input) returns ≥1 finding with evidence.pgEnumCount: 3
• evidence.tsUnionCount: 5 present in the finding
• evidence.enumName names the specific enum that drifted
• Mismatch does not produce empty findings[]

Forbidden: Mismatch exists but auditor returns empty findings[].

Evidence: unit — drmg/__tests__/story-s4-generator.spec.ts

Commission Result: ⬜ PASS / ⬜ FAIL Notes: (findings)

S5 — Virtue auditor proves the loop is improving​

Trigger: 3 prior audit result records exist in DB with increasing generator tier catches across runs 1→2→3

Checklist:

• auditVirtue(runs) returns finding where evidence.trend: "improving"
• evidence.periods contains data from all 3 runs
• Auditor with 0 or 1 run in DB produces no non-empty finding — requires ≥3 runs
• Flat or declining trend does not return "improving"

Forbidden: Virtue auditor runs with 0 or 1 run and returns a non-empty finding.

Evidence: integration — drmg/__tests__/story-s5-virtue.spec.ts

Commission Result: ⬜ PASS / ⬜ FAIL Notes: (findings)

Build Contract​

Success Test = the test in the Story Contract that goes GREEN when this row is done. Safety Test = what must NOT happen.

Context​

• PRD Index — Agent Platform
• Prompt Deck — 5-card pitch
• Pictures — Pre-flight maps

Questions​

When the virtue dimension shows a flat trend, is the problem the retrospectives or the routing?

• If a generator auditor finds zero issues, does that mean the generator is perfect — or that the auditor's input source is wrong?
• At what point does an 8-dimension audit become overhead rather than prevention?
• Which dimension catches the most findings in the first 5 runs — and does that reveal the weakest enforcement tier?