Skip to content

title: "Judge — "Is the quality sufficient?"" source: "tasks/TFW-46__evidence_layer/phase-a/review/judge.md"


Judge — "Is the quality sufficient?"

Mindset: Judge. You have the evidence from Verify. Now rule on quality. Every ✅ needs proof. Every ❌ needs a specific finding. Test: "Would I stake my reputation on this passing production review?" Mode: code Verify findings: verify.md

Universal Checklist

# Check Status Evidence
1 DoD met? All 6 ACs verified against actual files (verify.md V1-V12). AC-1 through AC-6 all match TS requirements. One minor stale ref found (RF.md line 68: §5 Observations should be §6) — cosmetic, doesn't affect AC fulfillment
2 Philosophy aligned P1 (real over synthetic): Evidence concept cleanly separates §4 synthetic from §5 real — verified in conventions.md §3 (V1). P2 (honest incompleteness): DEFERRED/BLOCKED/N/A in vocabulary — verified in RF template (V3). P3 (coordinator designs, executor collects): role pipeline in conventions.md + TS Evidence field + RF §5 — verified (V1, V2, V3). P4 (domain-agnostic): no code-specific examples in any template — verified (V1-V5). P5 (proportional): Evidence field supports N/A/empty — verified (V2). P7 (artifacts over claims): §14 anti-patterns enforce artifact references — verified (V1)
3 Tech debt documented RF §6 Observations has 3 items: (1) compilable_contract.md pre-existing REVIEW §5/§7 error, (2) glossary.md stale RF §7 ref (Phase C scope), (3) HL.md §9 Diagrams hint style note. All are real issues, properly scoped
4 Style & standards ⚠️ Overall good: naming follows conventions, Evidence Sections table follows D39 per-template pattern, anti-patterns follow existing format. One style issue: RF.md line 68 stale ref §5 Observations — inconsistent with §8 SI ref on line 69 which was correctly updated to §8. This indicates incomplete find-and-replace within the same instruction block
5 Observations collected 3 observations, all substantive: (1) pre-existing compilable_contract bug, (2) glossary stale ref correctly deferred to Phase C, (3) HL.md style note. Quality filter passes — these are real issues that would affect the next developer
6 RF completeness (§7-9) §7 Fact Candidates: "No fact candidates." — present, explicit N/A. §8 Strategic Insights: "No strategic insights." — present, explicit N/A. §9 Diagrams: "No diagrams." — present, explicit N/A. All sections present with explicit N/A per F21 pattern
7 Evidence completeness All 6 TS Evidence fields are N/A (conventions/template spec — validation is when users write real TS/RF with the new fields). RF §5 correctly mirrors this with 6/6 VERIFIED using local file references. Appropriate for a template-only task

Mode-Specific Checklist

# Check Status Evidence
7 Code quality Markdown templates follow existing conventions. Evidence Sections table in conventions.md §3 follows the same pattern as Visual Sections and Knowledge Capture tables (4 columns: Template, Section, Cognitive Mode, What it produces). Anti-patterns follow existing format. TS Evidence field instruction block is clear and concise
8 Test coverage N/A No executable tests for .tfw/ markdown. Executor ran 2 grep commands for reference validation
9 Security No secrets, no external access. Framework template changes only
10 Breaking changes ⚠️ RF section renumbering (§5-8 → §6-9) is a breaking change for any existing project RF files that reference old numbers. TS §8 Phase Risks acknowledges this: "Renumbering applies only to .tfw/ templates and workflows. Existing project RFs keep their numbering until next task." Mitigated by scope — only template/workflow files updated, not user project files

HL §7 Principles Check

# Principle Mapped AC AC met? Evidence
P1 Real over synthetic AC-1 conventions.md §3 clearly separates §4 (synthetic) from §5 (real) — V1
P2 Honest incompleteness AC-1, AC-3 Status vocabulary in conventions.md §3 and RF template §5 — V1, V3
P3 Coordinator designs, executor collects AC-2, AC-3 TS Evidence field + RF §5 Evidence section — V2, V3
P4 Domain-agnostic by default AC-1, AC-2 No domain-specific examples in any template — V1-V5
P5 Proportional to risk AC-2 Evidence field supports N/A, empty, DEFERRED — V2
P6 Tooling proactivity N/A (Phase B) TS §3 marks as Phase B scope
P7 Artifacts over claims AC-5 §14 anti-patterns require artifact references — V1

Contradictions with KNOWLEDGE.md

# Knowledge item RF claim Contradiction?
1 D39: Per-template visual sections Evidence Sections as new per-template table No — follows same pattern
2 F21: Explicit N/A pattern RF §7-9 all use "No X." pattern No — correctly applied
3 F13: Domain-agnostic Templates avoid code-specific examples No — correctly enforced

No contradictions found.

Checkpoint

Self-check: - [x] Every checklist item has evidence (not just ✅/❌)? - [x] Referenced verify.md findings in DoD assessment? - [x] Checked RF §7-9 for presence AND quality (not just existence)? - [x] KNOWLEDGE.md cross-referenced — contradictions documented or "None"? - [x] Fact Candidates from RF reviewed — any that need challenge?

"No fact candidates" challenge (Trust Protocol): This is a template-modification task with no human interaction during execution beyond the initial TS approval. No domain knowledge, stakeholder insights, or strategic decisions were surfaced. "No fact candidates" is legitimate.

Stage complete: YES