dew-method 0.1.0

package/src/dew-skills/3-solutioning/dew-transformation-design/steps/step-02-load-storage-ingestion-context.md

@@ -0,0 +1,65 @@
+# Step 02 — Load Storage and Ingestion Context
+
+## Mandatory Rules
+
+- Load storage targets and ingestion plans before transformation rules.
+- Do not invent Bronze/Silver/Gold targets.
+- Do not drop caveats.
+
+## Task
+
+Extract transformation-relevant context.
+
+## Extract from storage design
+
+- layer strategy
+- Bronze targets
+- Silver targets
+- Gold expectations
+- metadata requirements
+- schema snapshot strategy
+- quarantine targets
+
+## Extract from ingestion design
+
+- source ingestion pattern
+- ingestion frequency
+- schema evolution strategy
+- idempotency assumptions
+- replay/backfill limitations
+- metadata fields
+
+## Extract from KPI feasibility
+
+- verified/caveated KPIs
+- KPI formulas
+- KPI grain
+- required fields
+- caveats
+
+## Extract from source validation
+
+- source grain
+- source field coverage
+- quality caveats
+- freshness caveats
+
+## Update documents
+
+Update:
+- transformation-design.md
+- business-rule-catalog.md if known rules exist
+
+## Update frontmatter
+
+```yaml
+stepsCompleted:
+  - step-01-init
+  - step-02-load-storage-ingestion-context
+```
+
+## Continue checkpoint
+
+Ask:
+
+`[C] Continue to grain transformation gate`

package/src/dew-skills/3-solutioning/dew-transformation-design/steps/step-03-grain-transformation-gate.md

@@ -0,0 +1,65 @@
+# Step 03 — Grain Transformation Gate
+
+## Mandatory Rules
+
+- Stop at this gate.
+- Do not change grain silently.
+- User must approve grain transformation strategy.
+- If grain is undefined, trigger HALT-07.
+
+## Task
+
+Run Grain Transformation Gate.
+
+## Required table
+
+| Input Dataset | Input Grain | Output Dataset | Output Grain | Transformation Type | Risk |
+|---|---|---|---|---|---|
+
+Transformation types:
+- Preserve
+- Aggregate
+- Join / enrich
+- Bridge
+- Explode
+- Filter
+
+## Options
+
+A. Preserve source grain through Silver, aggregate only in Gold  
+B. Standardize to product/KPI grain in Silver  
+C. Use bridge/mapping tables between source grain and product grain  
+D. Keep multiple grains and document each model explicitly  
+E. Return to KPI/source validation because grain is unclear  
+
+## Recommendation
+
+Recommend one option.
+
+## User decision
+
+Ask user to choose A/B/C/D/E.
+
+Do not continue until user chooses.
+
+## Record
+
+Update:
+- transformation-design.md
+- transformation-decision-record.md
+- .decision-log.md
+- .learning-log.md if enabled
+
+Update frontmatter:
+
+```yaml
+decisionGates:
+  GATE-transformation-grain:
+    status: approved / approved-with-caveats / blocked
+```
+
+## Continue checkpoint
+
+Ask:
+
+`[C] Continue to Bronze to Silver rules`

package/src/dew-skills/3-solutioning/dew-transformation-design/steps/step-04-bronze-to-silver-rules.md

@@ -0,0 +1,65 @@
+# Step 04 — Bronze to Silver Rules
+
+## Mandatory Rules
+
+- Bronze to Silver rules must be source-specific.
+- Do not apply KPI-specific aggregation too early unless approved.
+- Metadata preservation must follow storage design.
+- DQ handoff must be explicit.
+
+## Task
+
+Define Bronze → Silver transformation rules.
+
+## For each source
+
+Record:
+
+| Rule ID | Source | Input Bronze | Output Silver | Rule Type | Description | Required? |
+|---|---|---|---|---|---|---|
+
+Rule types:
+- rename
+- cast
+- standardize
+- clean
+- deduplicate
+- filter
+- derive
+- join reference
+- validate
+- quarantine
+
+## Required sections
+
+For each Silver output:
+- purpose
+- grain
+- input Bronze dataset
+- output fields
+- metadata fields
+- transformation rules
+- DQ handoff
+- caveats
+
+## Update documents
+
+Update:
+- transformation-design.md
+- business-rule-catalog.md if business rules exist
+
+Update frontmatter:
+
+```yaml
+stepsCompleted:
+  - step-01-init
+  - step-02-load-storage-ingestion-context
+  - step-03-grain-transformation-gate
+  - step-04-bronze-to-silver-rules
+```
+
+## Continue checkpoint
+
+Ask:
+
+`[C] Continue to Silver to Gold rules`

package/src/dew-skills/3-solutioning/dew-transformation-design/steps/step-05-silver-to-gold-rules.md

@@ -0,0 +1,63 @@
+# Step 05 — Silver to Gold Rules
+
+## Mandatory Rules
+
+- Do not create Gold rule for unresolved KPI unless user approved caveat.
+- KPI formula must map to Gold fields.
+- Gold grain must be explicit.
+- Caveats must be preserved.
+
+## Task
+
+Define Silver → Gold transformation rules.
+
+## For each Gold output
+
+Record:
+- dataset name
+- consumer
+- KPI(s)
+- grain
+- input Silver datasets
+- transformation rules
+- KPI formula mapping
+- caveats
+
+## Required KPI mapping
+
+| KPI ID | Formula | Input Fields | Gold Field | Rule ID | Feasibility Status |
+|---|---|---|---|---|---|
+
+## If KPI is not verified/caveated
+
+Options:
+A. Exclude KPI from Gold  
+B. Include as caveated metric  
+C. Return to KPI feasibility  
+D. Replace KPI  
+
+Ask user if primary KPI is affected.
+
+## Update documents
+
+Update:
+- transformation-design.md
+- business-rule-catalog.md
+- transformation-decision-record.md if decisions occur
+
+Update frontmatter:
+
+```yaml
+stepsCompleted:
+  - step-01-init
+  - step-02-load-storage-ingestion-context
+  - step-03-grain-transformation-gate
+  - step-04-bronze-to-silver-rules
+  - step-05-silver-to-gold-rules
+```
+
+## Continue checkpoint
+
+Ask:
+
+`[C] Continue to business rule gate`

package/src/dew-skills/3-solutioning/dew-transformation-design/steps/step-06-business-rule-gate.md

@@ -0,0 +1,54 @@
+# Step 06 — Business Rule Gate
+
+## Mandatory Rules
+
+- Stop at this gate.
+- Business rules must be approved by user.
+- Business rules must have owner/change process.
+- Do not hide thresholds, mappings, or classification logic.
+
+## Task
+
+Run Transformation Rule / Business Rule Gate.
+
+## Required table
+
+| Rule ID | Rule Name | Rule Type | Related KPI | Owner | Change Process | Risk |
+|---|---|---|---|---|---|---|
+
+## Options
+
+A. Approve business rules as documented  
+B. Approve with caveats  
+C. Revise specific rules  
+D. Defer specific rules  
+E. Return to KPI/source validation  
+
+## User decision
+
+Ask user to choose A/B/C/D/E.
+
+Do not continue until user chooses.
+
+## Record
+
+Update:
+- business-rule-catalog.md
+- transformation-design.md
+- transformation-decision-record.md
+- .decision-log.md
+- .learning-log.md if enabled
+
+Update frontmatter:
+
+```yaml
+decisionGates:
+  GATE-19-transformation-rule:
+    status: approved / approved-with-caveats / blocked
+```
+
+## Continue checkpoint
+
+Ask:
+
+`[C] Continue to finalize transformation design`

package/src/dew-skills/3-solutioning/dew-transformation-design/steps/step-07-finalize.md

@@ -0,0 +1,42 @@
+# Step 07 — Finalize Transformation Design
+
+## Mandatory Rules
+
+- Do not finalize if grain gate is unresolved.
+- Do not finalize if business rule gate is unresolved.
+- Do not finalize if primary KPI Gold logic is unresolved unless user approved caveat.
+- Run checklist.
+
+## Task
+
+Finalize transformation design.
+
+## Required outputs
+
+- transformation-design.md
+- business-rule-catalog.md
+- transformation-decision-record.md
+- decision log entries
+- learning log entries if enabled
+
+## Final status
+
+Set:
+
+```yaml
+status: final
+stepsCompleted:
+  - step-01-init
+  - step-02-load-storage-ingestion-context
+  - step-03-grain-transformation-gate
+  - step-04-bronze-to-silver-rules
+  - step-05-silver-to-gold-rules
+  - step-06-business-rule-gate
+  - step-07-finalize
+```
+
+## Next workflow
+
+Recommend:
+
+`dew-data-model-design`

package/src/dew-skills/4-implementation/dew-code-review/SKILL.md

@@ -0,0 +1,44 @@
+---
+name: dew-code-review
+description: Review implemented data engineering story for AC compliance, DQ evidence, grain, lineage, operational behavior, and caveats.
+---
+
+# DEW Code Review
+
+**Goal:** Review an implemented DEW story before validation/done.
+
+**Your Role:** You are an adversarial data engineering reviewer.
+
+You do not approve code because it “looks okay”. You verify against story, design context, evidence, and data engineering risks.
+
+## Mandatory Rules
+
+- Review one story at a time.
+- Do not review story unless status is `review`.
+- Do not approve if required evidence is missing.
+- Do not approve if ACs are not satisfied.
+- Do not approve if DQ requirements are ignored.
+- Do not approve if code changes KPI formula, grain, source contract, or serving contract without approval.
+- Review must classify findings by severity.
+
+## Workflow Architecture
+
+This workflow uses step-file architecture.
+
+## On Activation
+
+1. Resolve customization.
+2. Load persistent facts.
+3. Load config if present.
+4. Load:
+   - `{workflow.code_review_report_template}`
+   - `{workflow.review_finding_rubric}`
+   - `{workflow.data_engineering_review_layers}`
+   - `{workflow.review_decision_rubric}`
+5. Greet user in configured language.
+
+## Execution
+
+Read fully and follow:
+
+`steps/step-01-load-story-and-changes.md`

package/src/dew-skills/4-implementation/dew-code-review/assets/code-review-report-template.md

@@ -0,0 +1,78 @@
+# Code Review Report Template
+
+```markdown
+---
+title: "Code Review — {Story ID}"
+status: draft
+created: {YYYY-MM-DD}
+updated: {YYYY-MM-DD}
+workflow: dew
+phase: "4-implementation"
+storyId: "{Story ID}"
+---
+
+# Code Review: {Story ID} — {Story Title}
+
+## 1. Review Summary
+
+Recommended decision:
+
+- approve
+- approve-with-caveats
+- request-changes
+- block
+
+## 2. Story Reviewed
+
+Story path:
+
+Status:
+
+## 3. Files Reviewed
+
+| Path | Purpose | Finding |
+|---|---|---|
+
+## 4. Acceptance Criteria Audit
+
+| AC | Status | Evidence | Notes |
+|---|---|---|---|
+| AC-1 | pass/fail/partial | | |
+
+## 5. Data Engineering Review
+
+| Layer | Status | Finding |
+|---|---|---|
+| Source access | | |
+| Ingestion | | |
+| Storage | | |
+| Transformation | | |
+| Model grain | | |
+| DQ | | |
+| Serving | | |
+| Governance/Security | | |
+| DataOps | | |
+
+## 6. Evidence Review
+
+| Required Evidence | Present? | Path | Valid? |
+|---|---|---|---|
+| | Yes/No | | Yes/No |
+
+## 7. Findings
+
+| Severity | Finding | Impact | Required Action |
+|---|---|---|---|
+| Critical/High/Medium/Low | | | |
+
+## 8. Decision
+
+A. Approve  
+B. Approve with caveats  
+C. Request changes  
+D. Block  
+
+User/reviewer decision:
+
+## 9. Required Follow-up
+```

package/src/dew-skills/4-implementation/dew-code-review/assets/data-engineering-review-layers.md

@@ -0,0 +1,61 @@
+# Data Engineering Review Layers
+
+## Acceptance Auditor
+
+Check:
+- each AC
+- implementation task completion
+- validation approach
+- story DoD
+
+## Source/Schema Hunter
+
+Check:
+- source path/API matches design
+- schema fields not invented
+- source contract respected
+- schema drift handling included
+
+## Grain Auditor
+
+Check:
+- model/output grain matches design
+- joins do not duplicate rows unexpectedly
+- aggregation rules are explicit
+
+## KPI Auditor
+
+Check:
+- KPI formula matches approved formula
+- KPI fields map correctly
+- output sample is plausible
+
+## DQ Auditor
+
+Check:
+- required DQ rules implemented
+- severity/action behavior respected
+- DQ evidence exists
+
+## Lineage/Metadata Auditor
+
+Check:
+- required metadata fields
+- source traceability
+- run IDs/timestamps
+- evidence paths
+
+## Operational Auditor
+
+Check:
+- logging
+- idempotency
+- failure handling
+- replay/backfill notes
+- run instructions
+
+## Caveat Auditor
+
+Check:
+- caveats from story/design are preserved
+- user-facing limitations not removed

package/src/dew-skills/4-implementation/dew-code-review/assets/review-decision-rubric.md

@@ -0,0 +1,33 @@
+# Review Decision Rubric
+
+## Approve
+
+Use when:
+- ACs pass
+- evidence exists
+- no critical/high findings
+- caveats documented
+
+## Approve with caveats
+
+Use when:
+- no critical findings
+- medium findings are acceptable
+- caveats are documented
+- validation still supports story goal
+
+## Request changes
+
+Use when:
+- high findings exist
+- missing non-critical evidence
+- AC partially met
+- correction is feasible
+
+## Block
+
+Use when:
+- critical finding exists
+- required evidence missing
+- story contradicts design
+- data output is unsafe/misleading

package/src/dew-skills/4-implementation/dew-code-review/assets/review-finding-rubric.md

@@ -0,0 +1,42 @@
+# Review Finding Rubric
+
+## Critical
+
+Blocks approval.
+
+Use when:
+- AC not met
+- required evidence missing
+- DQ critical rule ignored
+- KPI formula changed without approval
+- grain changed silently
+- sensitive/security issue introduced
+- output is misleading
+
+## High
+
+Usually blocks approval unless explicitly caveated.
+
+Use when:
+- partial evidence
+- important validation missing
+- idempotency/failure handling missing for pipeline story
+- source/schema caveat ignored
+
+## Medium
+
+Can approve with caveats.
+
+Use when:
+- documentation incomplete
+- minor operational issue
+- non-critical DQ evidence weak
+
+## Low
+
+Does not block.
+
+Use when:
+- naming
+- formatting
+- minor clarity issue

package/src/dew-skills/4-implementation/dew-code-review/checklist.md

@@ -0,0 +1,15 @@
+# Code Review Checklist
+
+- [ ] Story loaded and status is review.
+- [ ] Story context loaded.
+- [ ] Dev notes loaded.
+- [ ] Evidence manifest loaded.
+- [ ] File list reviewed.
+- [ ] ACs audited.
+- [ ] Required evidence reviewed.
+- [ ] DQ requirements reviewed.
+- [ ] Grain/KPI/source assumptions reviewed.
+- [ ] Metadata/lineage reviewed.
+- [ ] Operational requirements reviewed.
+- [ ] Findings classified.
+- [ ] Review decision recorded.

package/src/dew-skills/4-implementation/dew-code-review/customize.toml

@@ -0,0 +1,30 @@
+# DO NOT EDIT -- overwritten on every update.
+#
+# Workflow customization surface for dew-code-review.
+
+[workflow]
+
+activation_steps_prepend = []
+activation_steps_append = []
+
+persistent_facts = [
+  "file:{project-root}/**/stories/**/*.md",
+  "file:{project-root}/**/story-context/**/*.md",
+  "file:{project-root}/**/dev-notes/**/*.md",
+  "file:{project-root}/**/story-evidence/**/*.md",
+  "file:{project-root}/**/sprint-status.yaml",
+  "file:{project-root}/**/validation-evidence-index.md",
+  "file:{project-root}/**/.decision-log.md",
+  "file:{project-root}/**/.learning-log.md",
+]
+
+on_complete = ""
+
+code_review_report_template = "assets/code-review-report-template.md"
+review_finding_rubric = "assets/review-finding-rubric.md"
+data_engineering_review_layers = "assets/data-engineering-review-layers.md"
+review_decision_rubric = "assets/review-decision-rubric.md"
+checklist = "checklist.md"
+
+review_output_path = "{implementation_artifacts}/code-review"
+sprint_status_path = "{implementation_artifacts}/sprint/sprint-status.yaml"

package/src/dew-skills/4-implementation/dew-code-review/steps/step-01-load-story-and-changes.md

@@ -0,0 +1,32 @@
+# Step 01 — Load Story and Changes
+
+## Mandatory Rules
+
+- Review only story status `review`.
+- Load story, context, dev notes, evidence, file list.
+- Do not approve before full review.
+
+## Task
+
+Load review context.
+
+## Required inputs
+
+- story file
+- story context pack
+- dev notes
+- evidence manifest
+- file list
+- sprint-status.yaml
+
+## If missing
+
+- story not review → return to `dew-dev-story`
+- evidence missing → `HALT-18`
+- context missing → `HALT-17`
+
+## Continue checkpoint
+
+Ask:
+
+`[C] Continue to acceptance audit`