Eight Monday mornings, solved with one prompt each.

/sap-debug — MESSAGE_TYPE_X in VA01, plant 1000, customer 1234567, triggered on save. ST22 dump ID 20260422-0931-USR03.

What Claude does next

• Layer classification (ABAP / Basis / Functional / Integration) within 30 seconds
• SAP Note search auto-scoped by error + component (SD-SLS-SO)
• Dispatches sap-reviewer to pull the dump source and walk the stack
• Hard Gate: won't propose a fix until ST22 + SM21 evidence is attached
• Output: root-cause document with the fix, transport, and prevention note

# Root Cause — VA01 MESSAGE_TYPE_X (2026-04-22)

Layer:            ABAP · Authorization
Dump ID:          20260422-0931-USR03
Component:        SD-SLS-SO
Root cause:       AUTHORITY-CHECK OBJECT 'V_VBAK_VKO' failed.
                  User USR03 missing ACTVT=01 for sales org 1000.

Fix:              Add V_VBAK_VKO (VKORG=1000, ACTVT=01,02,03) to
                  role Z_SD_USER. Mass-assign via SU10.
Transport:        TRK900123 (DEV → QAS → PRD)
Related Notes:    2843117, 3092411
Evidence:         [x] ST22 dump attached  [x] SU53 trace attached
Prevention:       Add SoD check in Z_SD_USER review scheduled for next sprint.

Next skill →      transport-release

/sap-estimate — 3 new Fiori apps (display / approve / analytics) on PO release workflow. S/4HANA 2023 on-prem. Team: 2 mid-level ABAP devs, 1 senior Fiori, 1 functional. 6-week window.

What Claude does next

• WBS across ABAP, config, Basis, data migration, testing, cutover
• 6 complexity multipliers applied (new-dev vs enhancement, clean-core vs custom, ABAP vs CDS/RAP…)
• Three-point PERT range (optimistic · likely · pessimistic) with documented assumptions
• Risk buffer calculated per scenario, not hand-waved
• Hard Gate: refuses to emit a single-number estimate. The PMO sees the ranges.

Estimate · 3 Fiori apps on PO release (S/4HANA 2023)

WBS                            Base    ×Complexity    Effort
─────────────────────────────────────────────────────────────
CDS consumption view × 3       3d      × 1.4 (HANA)   4.2d
RAP behavior definition × 3    4d      × 1.6 (new)    6.4d
Fiori UI (3 apps)              9d      × 1.3 (std)   11.7d
Auth objects + PFCG roles      2d      × 1.2          2.4d
SIT + UAT packs                5d      × 1.1          5.5d
Cutover + transport discipline 2d      × 1.0          2.0d
─────────────────────────────────────────────────────────────
                                             Subtotal: 32.2d

PERT range (90% CI):
  Optimistic    28d · Team experienced, reqs locked
  Likely        34d · Two minor reqs drift; one integration iteration
  Pessimistic   44d · Reqs drift, BTP integration added late

Risk buffer:    +15%  (standard clean-core project)
Defensible #:   34-44d (2 scenarios)

Assumptions:
  · 2 mid ABAP + 1 senior Fiori + 1 functional, 80% allocation
  · OData via RAP (not legacy gateway)
  · No cross-tenant integration in scope
  · UAT lead-time: 5 days, not included in build

/sap-review — paste the ABAP below. Target: S/4HANA 2023, ABAP Cloud Tier 3. Flag BSEG reads, non-CDS extracts, and any clean-core violations.

What Claude does next

• Four-dimension review: quality, performance, security, clean core
• CRITICAL findings surface first (direct BSEG selects → CDS rewrites suggested inline)
• ATC-style remarks with SAP Note references for deprecated APIs
• Dispatches sap-reviewer for the deep scan, sap-security-auditor for auth-object risks
• Output: reviewable report you can paste into a merge request as-is

Code Review · ZME21N_RELEASE_ENHANCEMENT (412 lines)

CRITICAL (2)
  [L127] Direct SELECT on BSEG bypasses S/4HANA data model.
         Replace with CDS view I_JournalEntryItem.
         Performance delta: ~18× faster on 10M row tables.
         See Note 2185026.

  [L203] Native SQL (ADBC) — not ABAP-Cloud safe.
         Refactor to standard Open SQL or use CDS table function.

HIGH (3)
  [L88]  Hard-coded company code '1000'. Extract to constant or
         parameterise via release strategy table T16FG.
  [L245] Implicit enhancement point — avoid. Use BAdI BADI_PO_RELEASE.
  [L301] SY-SUBRC not checked after CALL FUNCTION 'BAPI_PO_CHANGE'.

MEDIUM (4)    [L15, L67, L158, L330] — naming, inline docs, unused vars

Clean-core score:  3 / 10  (critical violations block activation)
Security audit:    Auth object M_BEST_BSA check present ✓

Recommendation:    Do not merge. Fix CRITICAL + HIGH before release
                   window. Est. rework: 4h. sap-reviewer dispatched
                   for full line-by-line pass.

/sap-kickoff — greenfield S/4HANA Cloud public, retail client, Germany. Scope: FI, MM, SD. Go-live target: 10 months. Team size: 24 blended.

What Claude does next

• SAP Activate Discover + Prepare playbook sequenced
• Project charter with scope, governance, RACI, risk register
• Stakeholder map + role-based RACI matrix
• Industry overlay (retail) + country overlay (Germany: e-invoicing, DSGVO, DATEV)
• Chains into brainstorming → estimation → solution-architecture automatically

Project Charter · ACME DE Retail S/4HANA Implementation

Vision        Migrate from ECC 6.0 to S/4HANA Cloud public edition
              with FI + MM + SD in scope. Go-live 2026-12.

Scope         IN    3 company codes, 4 plants, 2 sales orgs, DATEV export
              IN    German e-invoicing (X-Rechnung via Ariba)
              OUT   HCM, PP, SuccessFactors (phase 2)
              OUT   In-store POS integration (vendor-owned)

Team (24)     Exec sponsor (client CFO) · Programme lead · 2 stream leads
              · 6 functional · 4 technical · 3 Basis · 2 data · Change + OCM

RACI          Decision-gate ownership matrix (R-A-C-I) for all 14
              Activate deliverables. Exec sign-off required for Gate 3 +
              Gate 7.

Risks         R-01  Master-data cleanse slippage (MM + SD)   · High
              R-02  DATEV SKR04 mapping gaps                 · High
              R-03  Custom Z-code compatibility (ECC → S/4)  · Medium
              R-04  German e-invoicing deadline 2026-09      · Medium
              R-05  Change resistance in Plant Bremen        · Medium

Phase dates   Discover    closed · Prepare 2026-04 → 2026-05
              Explore     2026-06 → 2026-08
              Realize     2026-08 → 2026-11
              Deploy      2026-11 → 2026-12

Next skill →  estimation · solution-architecture · change-management

/sap-doc FS — PO approval workflow. Dual approval above €50k (mgr + finance). Tolerance 5% on price, 10% on quantity. Release strategy in ME59N. Integrate with Ariba for vendor comms.

What Claude does next

• Structured FS: objective → scope → as-is → to-be → process steps → config → interfaces → authorizations
• BPMN-ready to-be flow with role swimlanes
• Config table references (T16FC, T16FK, T16FG) with real SPRO paths
• Test cases auto-sketched by sap-test-designer for the UAT pack
• Output: 15-25 page FS, reviewer-ready, in the KaarTech template shape

FS-MM-043 · Purchase Order Approval Workflow (excerpt)

1. OBJECTIVE
   Introduce tolerance-based release strategy for POs > €50k with
   dual approval (plant manager + finance controller).

3. TO-BE PROCESS FLOW (BPMN-ready)
   Buyer → [PO Draft] → ME21N → Release Strategy check
                                  │
                                  ├ ≤ €50k ───→ Auto-release
                                  │
                                  └ > €50k  ───→ Mgr approval (SAP
                                                 Inbox / Ariba Guided
                                                 Buying) → Finance
                                                 approval → Release
                                                 → Vendor dispatch

5. CONFIG (SPRO)
   SPRO → MM → Purchasing → PO → Release Procedure
     T16FC   Release codes       Z1 (Mgr), Z2 (Fin), Z3 (CFO>€250k)
     T16FK   Release strategies  Z1 → Z2 sequential, €50k-€250k
                                 Z1 → Z2 → Z3 sequential, > €250k
     T16FG   Release group       ZPO (all plants)
     T16FS   Classification      CEKKO class, chars NETWR, BUKRS
     CL01    Class ZPO_REL       tolerance ±5% price, ±10% qty

7. INTERFACES
     Ariba Guided Buying ↔ S/4 via CIG (Cloud Integration Gateway)
     Slack notifications via Ariba native connector (no custom dev)

9. TEST CASES (generated by sap-test-designer)
   TC-01  Standard PO €30k, 1 line, in tolerance  · Positive  · P1
   TC-02  PO €75k, mgr + fin dual approval        · Positive  · P1
   TC-03  PO €75k, fin rejects                    · Negative  · P2
   TC-04  PO €30k, price exceeds tolerance        · Negative  · P1
   TC-05  PO €300k, CFO signature required        · Positive  · P1
   (... 12 more ...)

Next skill →  code-generation (BAdI scaffolding) · testing-strategy

/sap-migrate — assess attached Z-code package for S/4HANA 2023 compatibility. Flag simplification items, BSEG reads, and obsolete function modules. Target: brownfield SUM migration.

What Claude does next

• Must-fix list separated from nice-to-fix
• Every BSEG read flagged with the ACDOCA-via-CDS rewrite
• Obsolete FM usage mapped to S/4 replacements
• Effort estimate per object, sorted by risk × effort
• Dispatches sap-migration-analyzer for the deep SUM compatibility scan

S/4HANA Migration Assessment · Z-package ZACME_MM (82,410 LOC)

MUST-FIX (43 objects · ~31 person-days)
  · Direct BSEG reads               12 objects   Note 1976487
    → Rewrite via CDS I_JournalEntryItem (simplification #1)
  · KNA1/LFA1 direct access          8 objects   Note 2265093
    → Use Business Partner (BP tcode). Backward flag BPRELATIONS.
  · MATDOC.NAME1 not in S/4          3 reports   Note 2161848
    → Redesign inventory reports on MATDOC.
  · BAPI_REQUISITION_CREATE obsolete 5 calls     Note 2928222
    → Migrate to BAPI_PR_CREATE.
  · Classical Dynpro + MODULE POOL  15 programs  Note 2226792
    → Fiori List Report / RAP Managed BO recommended.

NICE-TO-FIX (29 objects · ~14 person-days)
  · Native SQL (deprecated)          6 objects   Replace with Open SQL
  · SELECT * into TABLES             18 objects  Explicit field lists
  · Unused function group ZFG_OLD    5 programs  Archive

CLEAN (10 objects · passes as-is)

Effort heatmap (risk × effort):
              Low-effort    Mid-effort    High-effort
  High-risk   ─             [KNA1 calls]  [BSEG reads]
  Mid-risk    [Native SQL]  [Dynpros]     [BAPI swap]
  Low-risk    [SELECT *]    ─             ─

Recommended sequencing:
  Sprint 1  BSEG + KNA1 (high-risk, unblocks Gate 3)
  Sprint 2  BAPI swap + Dynpro decommission
  Sprint 3  Native SQL + SELECT * cleanup (parallel-safe)

Next skill →  estimation (effort finalised) · code-generation (CDS rewrites)

/sap-go-live-readiness — S/4HANA 2023, target cutover 2026-05-13. Scope: FI + MM + SD across 3 company codes. UAT 87% complete. Perf test pending.

What Claude does next

• 10-gate checklist — you literally cannot advance without evidence
• Required artefacts per gate (UAT sign-off, perf numbers, cutover dry-run log)
• Cutover runbook with explicit rollback triggers and decision owners
• Final Go / Conditional-Go / No-Go matrix the steering committee trusts
• Hard Gate: the model can't say 'PASS' without the attached artefact

Go-Live Readiness · Cutover 2026-05-13

Gate  Title                         Owner           Evidence            Status
─────────────────────────────────────────────────────────────────────────────────
 1    SIT + regression              QA Lead         UAT-pack-v2.pdf     PASS ✓
 2    UAT sign-off                  Business Lead   6/6 streams signed  PASS ✓
 3    Master data cleanse           Data Lead       97.2% reconciled    PASS ✓
 4    Performance at prod volume    Perf Lead       load-report.pdf     FAIL ✗
 5    Security + auth review        GRC Lead        SoD-scan.xlsx       PASS ✓
 6    Interfaces verified           Integration     12/12 endpoints     PASS ✓
 7    Cutover dry-run               PMO             dry-run-log.md      PASS ✓
 8    Training + hypercare plan     Change Lead     23/25 trained       COND ⚠
 9    Rollback procedure tested     Basis Lead      rollback-test.md    PASS ✓
10    Exec Go/No-Go                 Steering        signed charter      PENDING

BLOCKING: Gate 4 — Performance at prod volume.
  Observed:  Month-end close ran 4h 32m vs SLA of 2h 00m.
  Root cause: CDS view I_UniversalJournal not using the
              HANA column store optimiser (batch job SAPLSPAK).
  Action:    sap-reviewer dispatched. ETA fix: 3 days.

RECOMMENDATION: Conditional-Go subject to Gate 4 + Gate 8 closure
by 2026-05-08. Rollback runbook signed by Basis. If Gate 4 not
closed by 05-10, recommend No-Go and 2-week slip to 2026-05-27.

Decision required from: CFO + CIO + CTO by 2026-05-08 14:00 CET.

/sap-accelerate order-to-cash — retail client, Germany. Multi-channel (store + e-com + wholesale). Credit check via Credit Management, revenue recognition on delivery.

What Claude does next

• Complete O2C process flow, fit/gap matrix, and 30 test cases — pre-built
• Config guide with SPRO paths for pricing, credit, billing
• Industry overlay (retail omnichannel) adapts the base process
• Country overlay (Germany e-invoicing + DSGVO) merged in
• You adapt what's specific, not the 70% that's standard

O2C Accelerator · Retail · Germany (generated package)

INCLUDED (pre-built)
  ├ 01-process-flow.md        End-to-end O2C BPMN (store + e-com + wholesale)
  ├ 02-fit-gap-matrix.xlsx    46 process steps · 7 gaps identified
  ├ 03-config-guide.md        Pricing, credit mgmt, billing, revenue rec
  ├ 04-test-cases.xlsx        30 cases across happy path + edge + negative
  ├ 05-data-dictionary.md     KNA1, KNB1, KNVV, VBAK/P, VBRK/P mappings
  ├ 06-authorizations.md      13 PFCG roles scoped by channel
  └ 07-interfaces.md          Salesforce ↔ SAP, Ariba Commerce, Sana e-com

FIT-GAP HIGHLIGHTS (7 gaps)
  G-01  Store POS integration uses custom XML (vendor-locked)    Custom
  G-02  Multi-channel pricing precedence needs override table    Config
  G-03  Credit check on pre-orders not covered by standard       Custom
  G-04  Revenue recognition on delivery (IFRS-15) already fits   Standard ✓
  G-05  DSGVO customer-data-deletion workflow                    Custom
  G-06  X-Rechnung / ZUGFeRD invoice formats for B2B             Config
  G-07  Consignment stock via EWM (vs decentral WM)              Config

COUNTRY OVERLAY (Germany)
  · Tax code determination: 19% / 7% / 0% (MWSt) via condition technique
  · E-invoicing: X-Rechnung (public sector) + ZUGFeRD (B2B)
  · Banking: SEPA XML pain.001 for direct debit
  · DATEV export from SAP FI (SKR04 chart of accounts)

INDUSTRY OVERLAY (Retail omnichannel)
  · Store-to-DC replenishment (MRP-based)
  · Season inventory lifecycle (Fashion variant enabled)
  · Multi-channel pricing override (store > e-com > wholesale)
  · Consumer returns + RMA process via CRM integration

You adapt:  Custom gaps (G-01, G-03, G-05) + client-specific master data
Not built:  Standard pricing, billing, delivery, AR — already in the box

Next skill →  fit-gap-analysis (for G-01, G-03, G-05) · solution-architecture