@raishin/vanguard-frontier-agentic 2.9.0 → 2.11.0

package/skills/microsoft/d365-success-by-design-governance/references/official-sources.md

@@ -0,0 +1,51 @@
+# Official sources
+
+Use this reference only when you need source grounding for Dynamics 365 Success by Design, FastTrack implementation guidance, or fit-gap analysis behavior.
+
+## Microsoft Learn documentation
+
+Use these as starting points, not as proof of the user's live project state:
+
+- https://learn.microsoft.com/dynamics365/guidance/implementation-guide/success-by-design
+- https://learn.microsoft.com/dynamics365/guidance/implementation-guide/overview
+- https://learn.microsoft.com/dynamics365/guidance/implementation-guide/process-focused-solution-fit-to-standard-fit-gap-analysis
+- https://learn.microsoft.com/dynamics365/guidance/implementation-portal/conduct-solution-blueprint-review-workshop
+- https://learn.microsoft.com/dynamics365/guidance/implementation-guide/prepare-to-go-live
+- https://learn.microsoft.com/dynamics365/guidance/implementation-guide/prepare-go-live-checklist
+- https://learn.microsoft.com/dynamics365/guidance/fasttrack/go-live-workshops
+- https://learn.microsoft.com/dynamics365/guidance/fasttrack/implementation-workshops
+- https://learn.microsoft.com/training/paths/use-success-design/
+- https://learn.microsoft.com/dynamics365/guidance/implementation-guide/project-governance-conclusion
+
+## Grounding rule
+
+Official documentation explains Success by Design framework behavior and FastTrack program guidance. It does not prove the user's actual project phase, artifact completeness, SBR status, or go-live readiness posture. Prefer documented project artifacts (SBR outputs, fit-gap logs, implementation review findings, go-live checklist sign-offs) over inference.
+
+## Service facts (verified 2026-06-16)
+
+Success by Design phases:
+- The Success by Design framework organizes the Dynamics 365 implementation lifecycle into five methodology-agnostic phases: **Strategize**, **Initiate**, **Implement**, **Prepare**, and **Operate**.
+- **Strategize**: Discovery mode — gather and validate business requirements, finalize the high-level solution approach, define environment strategy and organizational strategy.
+- **Initiate**: Define all in-scope workstreams, update the project plan. The **Solution Blueprint Review (SBR)** is the mandatory starting review conducted at or before the end of this phase.
+- **Implement**: Build the solution per the agreed design and scope. **Implementation Reviews** address findings from the SBR; they cover data model, security, integration, ALM, and testing strategy.
+- **Prepare**: Solution built and tested. Final UAT, training, cutover plan, mock go-live, support model, deployment runbook. The **Go-live Readiness Review** is the mandatory final gate.
+- **Operate**: Solution is live. Goal is stabilization and handoff toward the next phase of digital transformation.
+
+Solution Blueprint Review (SBR):
+- The SBR is mandatory and serves as the starting point for all Success by Design reviews.
+- It covers: program strategy, test strategy, business process strategy, application strategy, data strategy, integration strategy, intelligence strategy, security strategy, ALM strategy, and environment and capacity strategy.
+- Findings from the SBR trigger Implementation Reviews for deeper dives into specific risk areas.
+
+Fit-to-standard and fit-gap:
+- **Fit-to-standard analysis**: Compare current processes against standard Dynamics 365 processes in the Business Process Catalog. Adopt standard configuration wherever possible.
+- **Fit-gap analysis**: Identify requirements that standard processes cannot fulfill. Evaluate complexity, cost, maintenance burden, and product roadmap alignment before approving any customization or extension.
+- Customizations that recreate legacy system behavior without justification are the leading cause of implementation cost overruns and upgrade risk.
+
+Go-live readiness review:
+- The Go-live Readiness Review is mandatory and conducted during the Prepare phase.
+- Required evidence: SIT completion and sign-off, UAT completion and sign-off, performance testing, data migration plan and dry-run results, cutover plan with go/no-go criteria, user training completion, license count confirmation, production support plan.
+- The review produces a go/no-go decision that must be documented and signed by project stakeholders and the implementation lead.
+
+Review implications:
+- Do not approve go-live from intent alone. Require SBR outputs, fit-gap analysis, implementation review findings with resolutions, go-live checklist completion, and explicit business owner sign-off.
+- Documentation cannot prove the user's actual project phase, artifact state, or implementation review history.

package/skills/microsoft/d365-success-by-design-governance/references/safety-checklist.md

@@ -0,0 +1,38 @@
+# Safety checklist
+
+Use this reference before any recommendation involving go-live approval, phase gate waiver, production deployment authorization, or compliance-impacting implementation decisions in Dynamics 365 projects following the Success by Design framework.
+
+## Non-negotiables
+
+- Never ask users to paste credentials, tenant IDs, environment URLs, client secrets, certificates, or customer personally identifiable information into chat.
+- Use documented project artifacts or sanitized user-provided evidence for current-state claims; otherwise use documentation and label the evidence level.
+- Do not invent phase gate artifact states, SBR finding counts, fit-gap analysis results, customization counts, or project timelines.
+- Require explicit human approval before recommending any production deployment, go-live authorization, or phase gate waiver.
+- Use current official Microsoft Learn documentation for Success by Design framework behavior and FastTrack guidance.
+- Keep recommendations least-change, reversible, and scoped to the domain in question.
+- Production deployment and go/no-go decisions are live-guard gated. Always escalate to the project sponsor and implementation lead with environment access before execution.
+
+## Stress checks
+
+- What Success by Design phase gates have been skipped or have no documented artifacts?
+- What Solution Blueprint Review findings remain open or without a triggered Implementation Review?
+- What customizations or extensions lack a documented fit-gap justification, cost estimate, or roadmap alignment review?
+- What go-live readiness checklist items are incomplete or missing stakeholder sign-off?
+- What rollback plan exists if the production go-live fails within the cutover window?
+- What audit evidence is missing that the project steering committee or FastTrack team would expect?
+- What hypercare and production support plan is in place for post-go-live stabilization?
+
+## Evidence labels
+
+Use `documented artifact`, `user-provided evidence`, `documentation-based`, or `inference`. Documentation alone never proves the user's actual project phase, SBR completion status, fit-gap discipline, or go-live readiness.
+
+## Live-guard gate
+
+The following actions require explicit human confirmation and are out of scope for automated execution:
+
+- Authorizing production go-live or issuing a formal go/no-go decision
+- Waiving a mandatory Success by Design phase gate or SBR workshop
+- Approving a customization or extension without documented fit-gap analysis and architectural review
+- Bypassing a FastTrack Implementation Review finding without documented mitigation
+- Authorizing production environment provisioning or deployable package promotion to production
+- Signing off on UAT, performance testing, or data migration dry-run results on behalf of business stakeholders

package/skills/microsoft/d365-success-by-design-governance/references/workflow-and-output.md

@@ -0,0 +1,70 @@
+# Workflow and output contract
+
+Use this reference only when performing the full Success by Design governance review, implementation phase gate assessment, or formatting the final answer.
+
+## Review domains
+
+Check these areas before giving a verdict:
+
+- Phase gate completeness: artifacts produced and reviewed per phase (Strategize, Initiate, Implement, Prepare, Operate)
+- Solution Blueprint Review: conducted, findings documented, Implementation Reviews triggered for high-risk areas
+- Fit-to-standard analysis: Business Process Catalog used, standard adoption documented, deviations justified
+- Fit-gap discipline: gaps logged, extension/customization decisions documented with risk and roadmap assessment
+- Customization sprawl: count and scope of extensions, ISV solutions, and custom code relative to standard product footprint
+- Implementation reviews: data model, security, integration, ALM, testing strategy reviews conducted where indicated by SBR
+- Go-live readiness: SIT, UAT, performance testing, data migration dry runs, cutover plan, support plan, license confirmation
+- Stakeholder sign-off: written go/no-go decision with named approvers
+- Post-go-live: stabilization plan, hypercare team readiness, lessons-learned review scheduled
+
+## Safe workflow
+
+1. **Frame scope**
+   - Implementation phase currently in scope:
+   - Dynamics 365 workloads (e.g., Finance, Supply Chain Management, Customer Service, Sales):
+   - FastTrack engagement: active / self-serve / none:
+   - Required outcome (phase gate review / SBR readiness / go-live approval / post-go-live):
+   - Explicit non-goals:
+
+2. **Collect evidence**
+   - Prefer documented artifacts: SBR output, fit-gap log, implementation review findings, go-live checklist, UAT sign-off, cutover plan.
+   - Otherwise inspect sanitized user-provided summaries or official Success by Design documentation.
+   - Label each finding as `documented artifact`, `user-provided evidence`, `documentation-based`, or `inference`.
+
+3. **Stress-test risk**
+   - What phase gates have been skipped or have missing artifacts?
+   - What SBR findings remain unresolved or without a triggered Implementation Review?
+   - What customizations lack a documented fit-gap justification or roadmap alignment assessment?
+   - What go-live readiness checklist items are incomplete or have no sign-off?
+   - What evidence is missing that would change the verdict?
+   - What rollback path exists if the go-live fails within the cutover window?
+
+4. **Recommend the smallest safe action**
+   - Prefer standard product adoption over customization, phased rollout over big-bang go-live, and SBR completion before implementation reviews.
+   - If the safest action is to stop and complete a missing phase gate (e.g., conduct the SBR before proceeding), say that plainly.
+   - Production deployment and go/no-go decisions require live-guard escalation. Do not recommend go-live without explicit stakeholder approval.
+
+## Output contract
+
+Return this structure:
+
+```markdown
+# D365 Success by Design Governance Review: <scope>
+## Executive verdict
+- Status: COMPLIANT / COMPLIANT WITH RISKS / NON-COMPLIANT / NEEDS EVIDENCE
+- Biggest risk:
+- Evidence level:
+## Scope and assumptions
+- Confirmed:
+- Unknown:
+- Out of scope:
+## Findings
+| Severity | Finding | Evidence | Why it matters | Minimum safe action |
+|---|---|---|---|---|
+## Recommended actions
+1. <action> — owner: <owner>, validation: <check>, rollback: <rollback>
+## Validation
+- Artifacts or checks to review:
+- Expected result:
+## Residual risk
+- <risk or explicit none>
+```

package/skills/microsoft/d365-supply-chain-plan-to-produce/SKILL.md

@@ -0,0 +1,59 @@
+---
+name: d365-supply-chain-plan-to-produce
+description: Review Dynamics 365 Supply Chain Management master planning (Planning Optimization/MRP), inventory management accuracy, procurement and sourcing configuration, warehouse management setup, and production control parameters including BOMs and routes. Enforces data-accuracy-before-planning discipline, detects coverage-settings and BOM configuration gaps, and requires live-guard escalation before production master-planning parameter or item-coverage changes. Refuses to approve planning output without inventory accuracy and coverage-settings evidence.
+allowed-tools: Read Grep Glob
+metadata:
+  author: "github: Raishin"
+  version: "0.1.0"
+  updated: "2026-06-16"
+  category: operational
+---
+
+# D365 Supply Chain Plan-to-Produce
+
+## Purpose
+
+Act as the Dynamics 365 Supply Chain Management planning and production reviewer who treats every unvalidated on-hand quantity, missing coverage setting, unapproved planned order, and unverified BOM or route as a production risk or supply disruption vector until evidenced otherwise.
+
+## When to use
+
+Use this skill for:
+
+- Master planning configuration review (Planning Optimization setup, master plan parameters, coverage groups, time fences, safety margins)
+- Inventory accuracy and coverage review (on-hand inventory validation, item coverage settings, safety stock levels, reorder points)
+- Procurement and sourcing configuration review (purchase policies, vendor agreements, lead times, sourcing rules)
+- Warehouse management parameter review (warehouse configuration, location directives, work templates, reservation hierarchies)
+- Production control setup review (production orders, BOMs, routes, operations scheduling, finite capacity settings)
+- Planned order review and firming discipline (planned purchase orders, planned production orders, planned transfer orders)
+- Demand forecasting and forecast model review (demand forecast setup, forecast reduction keys, forecast inclusion in master plans)
+- Intercompany master planning review (intercompany supply chain configuration, cross-legal-entity planned orders)
+- Supply risk and procurement efficiency analysis (supply risk assessment, vendor diversification, lead-time variability)
+- Audit evidence gathering for plan-to-produce compliance and operational controls
+
+## Lean operating rules
+
+- Prefer current Microsoft Learn documentation for Dynamics 365 Supply Chain Management master planning and production control service behavior. Use the per-skill facts and sources in `references/official-sources.md` for grounding.
+- Separate confirmed facts from inference. If state was not queried or shown, say so explicitly.
+- Challenge unvalidated on-hand quantities, missing safety stock definitions, unapproved planned order firm actions, coverage settings that create excess or stockout risk, and production parameter changes made without evidence.
+- Keep answers scoped, reversible, least-privilege, and explicit about blockers or unknowns.
+- Load references only when needed; do not pull all deep guidance into short answers.
+- Never ask for credentials, tenant IDs, environment URLs, connection strings, or customer supply chain data.
+
+## References
+
+Load these only when needed:
+
+- [Workflow and output contract](references/workflow-and-output.md) — use when executing the full master-planning or production-control review, or formatting the final answer.
+- [Safety checklist](references/safety-checklist.md) — use before any recommendation involving production master plan runs, coverage group reconfigurations, or BOM or route activations.
+- [Official sources](references/official-sources.md) — use when grounding Dynamics 365 Supply Chain Management master planning, inventory, or production control service behavior.
+- [Planning and production guide](references/planning-and-production-guide.md) — use for domain-specific failure modes, safe planning review workflow, verification targets, and pushback criteria.
+
+## Response minimum
+
+Return, at minimum:
+
+- the scoped target and evidence level,
+- the main planning gaps, inventory accuracy risks, or production-control deficiencies,
+- the safest next actions,
+- validation or rollback notes where relevant,
+- the assumptions or blockers that prevent stronger conclusions.

package/skills/microsoft/d365-supply-chain-plan-to-produce/metadata.json

@@ -0,0 +1,30 @@
+{
+  "id": "d365-supply-chain-plan-to-produce",
+  "name": "D365 Supply Chain Plan-to-Produce",
+  "type": "skill",
+  "provider": "microsoft",
+  "harnesses": [
+    "codex",
+    "claude-code",
+    "cursor",
+    "gemini",
+    "kiro",
+    "other"
+  ],
+  "summary": "Review Dynamics 365 Supply Chain Management master planning (Planning Optimization/MRP), inventory management accuracy, procurement and sourcing configuration, warehouse management setup, and production control parameters. Enforces data-accuracy-before-planning discipline, detects coverage-settings and BOM configuration gaps, and requires live-guard escalation before production master-planning parameter or item-coverage changes.",
+  "source_type": "original",
+  "official_docs": [
+    "https://learn.microsoft.com/dynamics365/supply-chain/master-planning/master-planning-home-page",
+    "https://learn.microsoft.com/dynamics365/supply-chain/supply-chain-management-welcome",
+    "https://learn.microsoft.com/training/modules/set-up-master-planning/",
+    "https://learn.microsoft.com/training/modules/use-master-planning/",
+    "https://learn.microsoft.com/dynamics365/supply-chain/production-control/production-process-overview"
+  ],
+  "security_notes": "Never approve master-planning parameter changes or item-coverage reconfigurations without inventory accuracy evidence and supply chain controller sign-off. Production master plan runs, coverage group modifications, and BOM or route activations are live-guard gated and must be escalated to a human supply chain manager or system administrator. Reject planning output approval where on-hand inventory data is unvalidated or safety stock levels are undocumented. Do not ask for credentials, tenant IDs, environment URLs, or customer supply chain data. Treat every unvalidated coverage setting, unapproved planned order firm action, and unreconciled inventory discrepancy as a production risk until evidenced otherwise. Procurement policy changes and vendor lead-time updates must be validated by a qualified supply chain manager before production use.",
+  "last_verified": "2026-06-16",
+  "path": "skills/microsoft/d365-supply-chain-plan-to-produce",
+  "author": "github: Raishin",
+  "version": "0.1.0",
+  "category": "operational",
+  "companion_agents": ["d365-supply-chain-plan-to-produce-agent"]
+}

package/skills/microsoft/d365-supply-chain-plan-to-produce/references/official-sources.md

@@ -0,0 +1,46 @@
+# Official sources
+
+Use this reference only when you need source grounding for Dynamics 365 Supply Chain Management master planning, inventory management, procurement, warehouse management, or production control service behavior, or the detailed source list.
+
+## Microsoft Learn documentation
+
+Use these as starting points, not as proof of the user's live environment state:
+
+- https://learn.microsoft.com/dynamics365/supply-chain/master-planning/master-planning-home-page
+- https://learn.microsoft.com/dynamics365/supply-chain/supply-chain-management-welcome
+- https://learn.microsoft.com/training/modules/set-up-master-planning/
+- https://learn.microsoft.com/training/modules/use-master-planning/
+- https://learn.microsoft.com/dynamics365/supply-chain/production-control/production-process-overview
+- https://learn.microsoft.com/dynamics365/supply-chain/inventory/inventory-home-page
+- https://learn.microsoft.com/dynamics365/supply-chain/procurement/procurement-sourcing-overview
+- https://learn.microsoft.com/dynamics365/supply-chain/warehousing/warehouse-configuration
+
+## Grounding rule
+
+Official documentation explains Dynamics 365 Supply Chain Management service behavior. It does not prove the user's current on-hand inventory positions, coverage settings, active master plans, planned order status, BOM versions, or route configurations. Prefer read-only evidence from the environment (e.g., planning log exports, on-hand inventory reports, coverage settings exports, production order status reports) over inference.
+
+## Service facts (verified 2026-06-16)
+
+Master planning model structure:
+- **Planning Optimization** is the current master planning engine for Dynamics 365 Supply Chain Management. The deprecated built-in master planning engine has been retired for new implementations.
+- The three main planning processes are: **Master planning** (net requirements, short-term), **Forecast planning** (gross requirements, long-term), and **Intercompany master planning** (net requirements across legal entities).
+- Planning Optimization runs outside the SQL database, minimizing impact on live operations and enabling runs during business hours with near-real-time results.
+- **Coverage groups** define replenishment settings per item: min/max, period, requirement, or safety stock coverage. Misconfigured coverage groups are the leading cause of excess inventory or stockout-driven production delays.
+
+Inventory and coverage:
+- **Safety stock** can be defined as a fixed quantity, a percentage of average demand, or a calculated value from demand forecasting. Undocumented safety stock levels create planning instability.
+- **Item coverage** overrides at the item-warehouse level take precedence over coverage group defaults. Unreviewed item-level overrides are a common source of planning anomalies.
+- **On-hand inventory** must be validated before trusting master planning output. Negative on-hand quantities, unposted inventory journals, and unclosed production orders all distort planning calculations.
+
+Procurement and sourcing:
+- **Purchase trade agreements** and **vendor default order settings** influence planned purchase order quantities, lead times, and vendor selection during planning runs. Stale trade agreement data produces unrealistic planned orders.
+- The **Supply risk assessment** feature (Dynamics 365 Supply Chain Management 10.0.31+) surfaces procurement risks for review before supply disruptions materialize.
+
+Production control:
+- **Bills of materials (BOMs)** and **routes** drive production order material and capacity requirements. Inactive or incorrect BOM versions and route operations produce inaccurate planned production orders.
+- **Operations scheduling** considers backlogged productions, material availability, and capacity availability. Finite capacity constraints are not considered in multi-threaded scheduling runs — document this limitation explicitly.
+- **Tracked components** (10.0.40+) enable batch and serial number registration for finished products and their components from the production floor execution interface.
+
+Review implications:
+- Do not approve master plan output or planned order firming without current on-hand inventory validation, coverage settings review, and supply chain manager sign-off.
+- Documentation cannot prove the user's actual inventory positions, active coverage settings, or production schedule state.

package/skills/microsoft/d365-supply-chain-plan-to-produce/references/planning-and-production-guide.md

@@ -0,0 +1,80 @@
+# Planning and production guide
+
+Use this reference for Dynamics 365 Supply Chain Management master planning and production control domain-specific failure modes, safe planning review workflow, verification targets, and pushback criteria.
+
+## What people get wrong
+
+The lazy story is:
+
+> Run Planning Optimization and firm the planned orders.
+
+Wrong. Planning Optimization output is only as reliable as its inputs: on-hand inventory accuracy, coverage settings, BOM and route validity, demand forecasts, and vendor lead time data. Firming incorrect planned orders without review creates purchase commitments, production orders, and transfer orders that cannot easily be reversed at scale.
+
+Common bad assumptions:
+
+- Planning Optimization output is always correct if the system runs without errors.
+- Coverage groups configured at go-live are still accurate after item portfolio or supplier changes.
+- Negative on-hand quantities in the system will be resolved by the next inventory count without impacting planning.
+- Auto-firming planned orders is safe because the system will only create what is needed.
+- BOM versions activated at go-live do not need periodic review against engineering change orders.
+- Safety stock levels set during implementation are still appropriate after demand pattern changes.
+
+## Planning failure modes
+
+- Coverage group settings not updated after item lead time changes, producing planned orders with incorrect replenishment timing.
+- On-hand inventory inflated by unposted inventory adjustment journals, causing Planning Optimization to under-suggest replenishment and creating production stockouts.
+- Negative on-hand quantities from inventory variances causing planning to calculate inflated requirements on dependent components.
+- Auto-firming horizon set too wide, firming planned orders before human review, creating incorrect purchase or production commitments.
+- BOM version with incorrect scrap percentages or component quantities causing planned production orders to under-order materials.
+- Safety margins (reorder margin, receipt margin, issue margin) not configured, causing planned orders to be scheduled without lead time buffers and arriving late.
+- Intercompany master planning not synchronized across legal entities, creating planned transfer orders that do not match demand signals.
+- Stale demand forecasts included in master plans after demand patterns shift, inflating planned supply beyond actual requirements.
+
+## High-risk planning control gaps (examples from plan-to-produce)
+
+- On-hand inventory not validated before the planning run (distorted net requirements produce incorrect planned order quantities)
+- Coverage type set to Period for fast-moving items (batches all demand into single large orders, creating cash flow and warehouse capacity risk)
+- Planned orders auto-firmed within the firming horizon without planner review (incorrect commitments that require manual cancellation)
+- BOM version inactive or referencing obsolete components (planned production orders cannot be released to production without BOM correction)
+- Safety stock set to zero for single-source critical components (no buffer against supplier delays)
+- Warehouse location directives not configured for new item locations (warehouse management work cannot be generated, blocking receipt or pick)
+- Procurement policy requiring purchase order approval bypassed during rush procurement (unauthorized spend commitments)
+
+These gaps represent the highest-risk plan-to-produce scenarios per supply chain operations and internal control guidance. Verify that each gap is addressed before approving planning output or production schedule reliability.
+
+## Minimum safe planning review workflow
+
+1. Confirm scope: legal entities, warehouses, item groups, and planning horizon in review.
+2. Validate on-hand inventory: run on-hand inventory report, confirm no unposted adjustment journals, resolve negative on-hand quantities.
+3. Review coverage settings: confirm coverage groups are current for the item portfolio; review item-level overrides for anomalies.
+4. Review safety stock levels: confirm documented and appropriate for current demand patterns and supplier lead times.
+5. Review BOM and route versions: confirm active versions are current, component quantities are correct, and routes match production capacity.
+6. Run Planning Optimization and review the planning log for warnings, errors, and action messages.
+7. Review planned orders within the firming horizon: confirm quantities, dates, and vendor or production resource assignments are reasonable.
+8. Confirm auto-firming settings: verify the auto-firming horizon does not exceed the safe review window for planners.
+9. For procurement: confirm vendor lead times in trade agreements are current before approving planned purchase orders.
+10. Require live-guard escalation for any production configuration change before recommending it.
+
+## Verification targets
+
+- On-hand inventory: validated, no unposted journals, no unexplained negative quantities
+- Coverage groups: current, documented, reviewed after item or supplier changes
+- Safety stock: documented levels per item, appropriate for lead time variability
+- BOM versions: active versions current, component quantities and scrap percentages validated
+- Route configurations: operations sequences current, finite capacity constraints documented
+- Planning log: reviewed for action messages, warnings, and errors
+- Planned order review: planner sign-off on firmed orders within the firming horizon
+- Auto-firming horizon: set within the safe planner review window
+
+## When to push back
+
+Push back if the user asks to:
+
+- approve master planning output without validating on-hand inventory accuracy first
+- accept negative on-hand quantities as normal without investigating root cause
+- firm planned orders in bulk without planner review of quantities, dates, and vendors
+- extend the auto-firming horizon beyond the planner's practical review capacity
+- approve a master plan run without confirming BOM and route versions are current and active
+- set safety stock to zero for critical or single-source items without documented risk acceptance
+- make production coverage group or master plan parameter changes without live-guard escalation and explicit human approval
+- trust planning output from a system where demand forecasts are stale or coverage groups have not been reviewed since go-live

package/skills/microsoft/d365-supply-chain-plan-to-produce/references/safety-checklist.md

@@ -0,0 +1,38 @@
+# Safety checklist
+
+Use this reference before any recommendation involving production master plan runs, coverage group reconfigurations, planned order firming, BOM or route activations, warehouse management parameter changes, or compliance-impacting supply chain configuration changes in Dynamics 365 Supply Chain Management.
+
+## Non-negotiables
+
+- Never ask users to paste credentials, tenant IDs, environment URLs, client secrets, certificates, or customer personally identifiable supply chain data into chat.
+- Use exported planning logs, on-hand inventory reports, or sanitized user-provided evidence for live-state claims; otherwise use documentation and label the evidence level.
+- Do not invent on-hand quantities, coverage settings, BOM version numbers, planned order quantities, lead times, or live environment state.
+- Require explicit human approval before recommending any production master plan parameter change, coverage group reconfiguration, planned order mass firming, or BOM or route activation.
+- Use current official Microsoft Learn documentation for Dynamics 365 Supply Chain Management master planning and production control behavior.
+- Keep recommendations least-disruptive, reversible where possible, and scoped to the domain in question.
+- Production master plan runs, coverage group changes, and BOM or route activations are live-guard gated. Always escalate to a qualified Dynamics 365 Supply Chain Management controller or system administrator with environment access before execution.
+
+## Stress checks
+
+- What items have missing or incorrect coverage settings that could create stockout or excess inventory upon the next planning run?
+- What on-hand inventory records are negative, unvalidated, or based on unposted journals that would distort planning output?
+- What planned orders are near the firming horizon without human review or approval, risking auto-firming of incorrect quantities?
+- What BOM versions or route operations are inactive, incorrectly configured, or based on stale data that would create production errors?
+- What rollback path exists if a master plan run produces incorrect planned orders that are auto-firmed before review?
+- What audit evidence is missing that supply chain managers or operations auditors would require?
+
+## Evidence labels
+
+Use `live evidence`, `report evidence`, `user-provided evidence`, `documentation-based`, or `inference`. Documentation alone never proves the user's live Dynamics 365 Supply Chain Management inventory positions, coverage settings, active BOM versions, or production schedule state.
+
+## Live-guard gate
+
+The following actions require explicit human confirmation and are out of scope for automated execution:
+
+- Running or canceling a master plan in any production environment
+- Modifying master plan parameters or coverage group settings in a production environment
+- Activating or deactivating BOM versions or route configurations in a production environment
+- Mass-firming planned orders in a production environment
+- Modifying warehouse configuration, location directives, or work templates in a production environment
+- Changing item coverage settings or safety stock levels in a production environment
+- Modifying procurement policies, purchase trade agreements, or vendor sourcing rules in a production environment

package/skills/microsoft/d365-supply-chain-plan-to-produce/references/workflow-and-output.md

@@ -0,0 +1,71 @@
+# Workflow and output contract
+
+Use this reference only when performing the full master-planning or production-control review, implementation guidance, operational evidence gathering, or formatting the final answer.
+
+## Review domains
+
+Check these areas before giving a verdict:
+
+- Master plan configuration: Planning Optimization setup, master plan parameters, coverage time fences, safety margins, action message settings
+- Coverage group and item coverage settings: coverage type, min/max quantities, safety stock levels, reorder points, lead time settings
+- Inventory accuracy: on-hand validation, negative on-hand quantities, unposted inventory journals, unclosed production orders
+- Procurement configuration: purchase trade agreements, vendor default order settings, sourcing rules, procurement policies, lead time accuracy
+- Warehouse management parameters: warehouse configuration, location directives, work template setup, reservation hierarchy accuracy
+- BOM and route configuration: active BOM versions, route operations accuracy, scrap percentages, finite capacity settings
+- Planned order review: planned purchase orders, planned production orders, planned transfer orders, auto-firming rules
+- Demand forecasting: forecast model setup, forecast reduction keys, forecast inclusion in master plans
+- Intercompany planning: cross-legal-entity supply chain configuration, intercompany planned order accuracy
+- Compensating controls: inventory count schedules, cycle count frequency, approval workflows for order firming
+
+## Safe workflow
+
+1. **Frame scope**
+   - Legal entities and warehouses in scope:
+   - Planning horizon (short-term, medium-term, long-term):
+   - Operational driver (inventory reduction, stockout prevention, production schedule reliability, procurement efficiency):
+   - Required outcome (planning config review / inventory accuracy review / production control audit):
+   - Explicit non-goals:
+
+2. **Collect evidence**
+   - Prefer exported planning logs, on-hand inventory reports, coverage settings exports, and production order status reports for current-state claims.
+   - Otherwise inspect sanitized user-provided evidence, configuration screenshots, or official Dynamics 365 Supply Chain Management documentation.
+   - Label each finding as `live evidence`, `report evidence`, `user-provided evidence`, `documentation-based`, or `inference`.
+
+3. **Stress-test risk**
+   - What items have missing or incorrect coverage settings that could cause stockout or excess inventory?
+   - What on-hand inventory records are negative, unvalidated, or based on unposted journals?
+   - What planned orders are near the firming horizon without review or approval?
+   - What BOM versions or route operations are inactive or incorrectly configured?
+   - What evidence is missing that would change the planning output verdict?
+   - What procurement lead times or vendor agreements are stale or unvalidated?
+
+4. **Recommend the smallest safe action**
+   - Prefer inventory validation, coverage-settings review, and planned-order analysis before firming production or purchase orders.
+   - If the safest action is to stop and gather evidence (run an on-hand inventory report or planning log first), say that plainly.
+   - Production master plan runs, coverage group reconfigurations, and BOM or route activations require live-guard escalation. Do not recommend live changes without explicit human approval.
+
+## Output contract
+
+Return this structure:
+
+```markdown
+# D365 Supply Chain Plan-to-Produce Review: <scope>
+## Executive verdict
+- Status: PLAN RELIABLE / PLAN WITH CONDITIONS / PLAN NOT RELIABLE / NEEDS EVIDENCE
+- Biggest risk:
+- Evidence level:
+## Scope and assumptions
+- Confirmed:
+- Unknown:
+- Out of scope:
+## Findings
+| Severity | Finding | Evidence | Why it matters | Minimum safe action |
+|---|---|---|---|---|
+## Recommended actions
+1. <action> — owner: <owner>, validation: <check>, rollback: <rollback>
+## Validation
+- Reports or checks to run:
+- Expected result:
+## Residual risk
+- <risk or explicit none>
+```

package/skills/microsoft/fabric-analytics-engineering/SKILL.md

@@ -0,0 +1,60 @@
+---
+name: fabric-analytics-engineering
+description: Review Microsoft Fabric analytics engineering artifacts — Fabric Data Warehouse T-SQL design and anti-patterns, dimensional modeling (star schema, fact and dimension tables, relationships), semantic model design (Direct Lake vs Import vs DirectQuery selection, table layout, relationship cardinality), DAX measure correctness and optimization (iterators, filter context, variables, CALCULATE), data preparation and transformation quality, and reusable certified semantic models feeding Power BI reports. Use to fix bad star schemas, slow DAX, untrustworthy measures, and warehouse anti-patterns. Distinct from governance: this is build quality and modeling correctness, not RLS or workspace trust. Static review only; production warehouse schema changes and semantic-model deployment are escalated.
+allowed-tools: Read Grep Glob
+metadata:
+  author: "github: Raishin"
+  version: "0.1.0"
+  updated: "2026-06-17"
+  category: data
+---
+
+# Fabric Analytics Engineering
+
+## Purpose
+
+Act as the Microsoft Fabric analytics engineering reviewer who treats every denormalized fact table, ambiguous many-to-many relationship, incorrect DAX filter context, calculated column used instead of an explicit measure, and DirectQuery fallback on a Direct Lake SQL view as a correctness and performance risk until proven otherwise.
+
+## When to use
+
+Use this skill for:
+
+- Fabric Data Warehouse T-SQL design: table definitions, star and snowflake schema correctness, fact and dimension table design, surrogate key patterns, SCD handling, multi-table ACID transaction patterns, materialized views, stored procedures, and T-SQL anti-patterns (implicit conversions, missing column statistics)
+- Dimensional modeling: star-schema design, fact table granularity, dimension key integrity, bridge tables for many-to-many, role-playing dimensions, degenerate dimensions, and slowly changing dimension strategies
+- Semantic model design: storage mode selection (Direct Lake on OneLake vs Direct Lake on SQL vs Import vs DirectQuery), relationship cardinality and cross-filter direction, composite models, large semantic model storage format, incremental refresh configuration, framing behavior and V-Order dependency
+- DAX quality: explicit measure correctness, filter context and row context separation, CALCULATE and FILTER usage, iterator functions (SUMX, AVERAGEX), calculation groups, dynamic format strings, field parameters, DAX variables for readability, and common anti-patterns (implicit measures, CALCULATE with ALL, ambiguous relationships)
+- Data preparation and transformation: T-SQL views and stored procedures as transformation layer, Dataflows Gen2 for analytics prep, star-schema loading patterns via pipelines, deduplication and null handling
+- Reusable certified semantic models: shared model design for Power BI reuse, endorsed/certified model patterns (Build permission, single source of truth) as a build-quality concern distinct from governance policy
+
+Do not use this skill for:
+
+- RLS/OLS enforcement, workspace role assignments, sensitivity labels, or DLP policies (use fabric-power-bi-business-insights-governance)
+- Lakehouse/Spark pipeline engineering, medallion architecture, CU capacity tuning, or Real-Time Intelligence (use fabric-data-engineering)
+- Power Platform or Dataverse modeling (use an appropriate Power Platform skill)
+
+## Lean operating rules
+
+- Prefer current Microsoft Learn documentation for Fabric Data Warehouse T-SQL behavior, Direct Lake semantics, DAX evaluation, and DP-600 skill areas.
+- Separate confirmed facts from inference. If warehouse DDL, semantic model metadata, or DAX measures were not provided, say so.
+- Challenge denormalized fact tables, missing surrogate keys, fan-out joins, bidirectional cross-filter on multiple paths, implicit measures, DirectQuery fallback caused by SQL views in Direct Lake models, and DAX that relies on calculated columns for aggregation.
+- Promote star-schema discipline: numeric surrogate keys, date dimension, single-direction relationships, explicit DAX measures, certified shared semantic models.
+- Keep answers scoped, reversible, and explicit about blockers or unknowns. Never ask for credentials, tenant IDs, workspace URLs, or customer data.
+- Load references only when needed.
+
+## References
+
+Load these only when needed:
+
+- [Workflow and output contract](references/workflow-and-output.md) — use when executing the full Fabric analytics engineering review or formatting the final answer.
+- [Safety checklist](references/safety-checklist.md) — use before any recommendation involving production warehouse schema changes, semantic-model deployment, or deployment-pipeline promotion.
+- [Official sources](references/official-sources.md) — use when grounding Fabric Data Warehouse T-SQL, dimensional modeling, Direct Lake, or DAX behavior.
+
+## Response minimum
+
+Return, at minimum:
+
+- the scoped target and evidence level,
+- the main dimensional modeling, T-SQL design, DAX correctness, or semantic-model configuration gaps,
+- the safest next actions,
+- validation or rollback notes where relevant,
+- the assumptions or blockers that prevent stronger conclusions.

package/skills/microsoft/fabric-analytics-engineering/metadata.json

@@ -0,0 +1,30 @@
+{
+  "id": "fabric-analytics-engineering",
+  "name": "Fabric Analytics Engineering",
+  "type": "skill",
+  "provider": "microsoft",
+  "harnesses": [
+    "codex",
+    "claude-code",
+    "cursor",
+    "gemini",
+    "kiro",
+    "other"
+  ],
+  "summary": "Review Microsoft Fabric analytics engineering artifacts: Fabric Data Warehouse T-SQL design and anti-patterns, dimensional modeling (star schema, fact and dimension tables, relationships), semantic model design (Direct Lake vs Import vs DirectQuery selection, table layout, relationship cardinality), DAX measure correctness and optimization (iterators, filter context, variables, CALCULATE), data preparation and transformation quality, and reusable certified semantic models feeding Power BI reports. Use to fix bad star schemas, slow DAX, untrustworthy measures, and warehouse anti-patterns. Distinct from governance: this skill covers build quality and modeling correctness, not RLS or workspace trust. Static review only; production warehouse schema changes and semantic-model deployment are live-guard gated.",
+  "source_type": "original",
+  "official_docs": [
+    "https://learn.microsoft.com/fabric/data-warehouse/dimensional-modeling-overview",
+    "https://learn.microsoft.com/fabric/fundamentals/direct-lake-overview",
+    "https://learn.microsoft.com/fabric/data-warehouse/data-warehousing",
+    "https://learn.microsoft.com/dax/dax-overview",
+    "https://learn.microsoft.com/credentials/certifications/resources/study-guides/dp-600"
+  ],
+  "security_notes": "Static review only. Never execute DDL or DML against production Fabric Data Warehouse schemas, deploy or overwrite production semantic models, publish Power BI reports to production workspaces, or modify deployment-pipeline stages without explicit human approval, blast-radius assessment, and a rollback path; these are live-guard gated and must be escalated to a Fabric or analytics administrator. Do not ask for credentials, tenant IDs, workspace URLs, or customer data. Treat missing surrogate keys in fact tables, fan-out or chasm-trap join patterns, incorrect DAX filter context, measures using calculated columns instead of explicit DAX, DirectQuery fallback on Direct Lake SQL views, and unverified measure outputs as correctness risks until reviewed. Label every finding as documented artifact, user-provided sanitized evidence, documentation-based, or inference.",
+  "last_verified": "2026-06-17",
+  "path": "skills/microsoft/fabric-analytics-engineering",
+  "author": "github: Raishin",
+  "version": "0.1.0",
+  "category": "data",
+  "companion_agents": ["fabric-analytics-engineering-agent"]
+}

package/skills/microsoft/fabric-analytics-engineering/references/official-sources.md

@@ -0,0 +1,17 @@
+# Official sources
+
+Use this reference only when you need source grounding for Microsoft Fabric analytics engineering, dimensional modeling, Direct Lake, DAX, or Fabric Data Warehouse T-SQL behavior.
+
+## Microsoft Learn documentation
+
+Use these as starting points, not as proof of the user's live warehouse schema, semantic model configuration, or DAX measure output:
+
+- https://learn.microsoft.com/fabric/data-warehouse/dimensional-modeling-overview — Dimensional modeling in Fabric Data Warehouse: star schema design (fact tables, dimension tables, surrogate keys), snowflake vs star trade-offs, slowly changing dimension strategies, bridge tables for many-to-many, load patterns, and T-SQL best practices for Fabric Warehouse. Supports schema design and dimensional-modeling review steps.
+- https://learn.microsoft.com/fabric/fundamentals/direct-lake-overview — Direct Lake overview: how framing works (metadata refresh vs full import), DirectQuery fallback triggers (SQL views, unsupported features), V-Order dependency for performance, Direct Lake on OneLake vs Direct Lake on SQL endpoints, composite model support, and framing latency considerations. Supports storage-mode selection and Direct Lake configuration review steps.
+- https://learn.microsoft.com/fabric/data-warehouse/data-warehousing — Fabric Data Warehouse overview: enterprise T-SQL warehouse on Delta Lake foundation, star/snowflake schema use cases, multi-table ACID transactions, materialized views, stored procedures, COPY INTO, cross-database queries, autonomous workload management, and Power BI integration. Supports warehouse design and anti-pattern review steps.
+- https://learn.microsoft.com/dax/dax-overview — DAX overview: measures vs calculated columns vs calculated tables, filter context and row context, CALCULATE, iterator functions (SUMX, AVERAGEX), evaluation context, time-intelligence functions, and common correctness pitfalls. Supports DAX measure quality and optimization review steps.
+- https://learn.microsoft.com/credentials/certifications/resources/study-guides/dp-600 — DP-600 Fabric Analytics Engineer Associate study guide: skill areas covering semantic model design (storage mode, star schema, relationships, DAX, calculation groups, composite models, incremental refresh), data preparation (T-SQL views, Dataflows Gen2, star-schema loading), and analytics lifecycle (deployment pipelines, XMLA endpoint, shared semantic models). Supports scope-framing and gap-identification steps.
+
+## Grounding rule
+
+Official documentation explains Fabric Data Warehouse, Direct Lake, and DAX behavior. It does not prove the user's actual warehouse schema, semantic model relationships, measure correctness, or DirectQuery fallback state. Prefer user-provided sanitized DDL, semantic model metadata (PBIP/TMDL), or DAX measure definitions for current-state claims. Label each finding as `documented artifact`, `user-provided evidence`, `documentation-based`, or `inference`.