feat(skills): add ABAP-to-BTP-CAP modernization skill chain (MVP)

[Raistlin82]

Summary

Adds a side-by-side ABAP → BTP CAP modernization skill chain to skills/. Takes a custom ABAP package (Z*) on S/4HANA / ECC and produces a complete BTP-native CAP project scaffold (CDS schema + service + handler stubs + future Fiori Elements V4 + MTA deployment artifacts) without touching the source ABAP system.

This is the MVP foundation (draft for review). Orchestrator + 2 core sub-skills included. Remaining 4 sub-skills tracked as follow-up — see Follow-up below.

Motivation

Existing arc-1 skills cover:

• In-place ABAP cloud readiness (migrate-custom-code) — fixes ATC findings inside the SAP system
• Clean Core classification (sap-clean-core-atc) — bucket Z code into Levels A-D
• In-stack RAP generation (generate-rap-service*) — new services on ABAP Cloud

But there's a gap: side-by-side BTP CAP modernization — when the target architecture is a CAP application on BTP (not ABAP Cloud), the workflow is different:

• Source ABAP stays read-only
• Target is a parallel BTP project (CDS + Fiori Elements + CF)
• Released S/4 APIs consumed remotely (no direct ABAP execution in target)
• Customer needs scaffold + plan to evaluate / commit to the BTP roadmap

This skill chain addresses that use case.

What's included

New skill files (3 markdown, ~1211 lines)

Skill	Lines	What it does
modernize-abap-to-btp-cap	357	End-to-end orchestrator: inventory → schema → service → fiori → auth → mta → ADR. Smart Defaults, BTP vs On-Prem comparison, error handling.
modernize-abap-cap-schema	370	DDIC → CDS type mapping (24 types). FK → Association/Composition inference. cuid / managed aspect auto-application. Currency / Quantity semantic annotations.
modernize-abap-cap-service	484	FM/program/class → CAP service action mapping. Bound vs unbound action classification. TypeScript handler stubs with TODO + ABAP source excerpts. 13 common ABAP → CAP handler pattern mappings.

README update

New section "BTP CAP Modernization (Preview)" in skills/README.md:

• Differentiates from sap-clean-core-atc / migrate-custom-code (those keep source in SAP)
• Documents 7-step typical workflow
• Cross-references existing skills (bootstrap-system-context, sap-unused-code, sap-clean-core-atc, generate-cds-unit-test)

Design decisions

• Side-by-side, NOT in-place: leaves source ABAP system read-only. Generates parallel CAP target. NO SAPWrite calls needed.
• Skill-first contribution: aligned with arc-1's 17-skill community pattern. NO new MCP tool, NO core code change. Honors the 12-tool token-efficient principle (CLAUDE.md §Design Principles).
• Markdown orchestration: skills chain via inline markdown links (consistent with sap-clean-core-atc → migrate-custom-code pattern).
• Sandbox output by default: writes to <target>/.target-cap-staging/ for review before apply. Reversible.
• Smart Defaults: every decision documented with rationale; avoid asking the user unless ambiguous.
• TODO scaffolds, NOT auto-implementation: handler bodies are stubs with embedded ABAP source excerpts; user implements business logic in a second pass. Honest about what the LLM can/cannot translate.

Use cases addressed

• ✅ Greenfield BTP CAP migration of a Z* package from ECC / on-prem S/4
• ✅ Lift-and-redesign for BTP-native architecture
• ✅ POC / pilot for evaluating BTP CAP as target
• ✅ Quarterly modernization assessment (run with --skip flags for analysis-only)
• ✅ Pre-deployment audit before committing to BTP roadmap

What this skill chain does NOT do (intentionally out of scope)

• ❌ No SAP write operations
• ❌ No CF deployment automation (cf push stays manual)
• ❌ No HANA service provisioning
• ❌ No XSUAA service creation
• ❌ No automatic handler logic generation (TODO stubs only)
• ❌ No regression test execution
• ❌ No transport-coordinated migration
• ❌ No Kyma target (CF only for v1)
• ❌ No CAP Java target (Node.js only for v1)
• ❌ No Freestyle UI5 (Fiori Elements V4 only)

Follow-up

The orchestrator references 4 additional sub-skills that are not yet included in this PR (tracked as next steps):

• modernize-abap-clean-core-gap (~400 lines) — Batch Clean Core L-A per-edition compliance check + replacement suggestions
• modernize-abap-fiori-elements (~500 lines) — Service → Fiori Elements V4 LROP scaffold (manifest + annotations)
• modernize-abap-auth-mapping (~300 lines) — AUTHORITY-CHECK → XSUAA scopes + @restrict annotations
• modernize-abap-btp-mta (~300 lines) — CAP project → mta.yaml + Dockerfile + manifest.yml

The orchestrator handles missing sub-skills gracefully (skip with warning). Sub-skills will be added in follow-up PRs once the MVP gets review feedback.

Compatibility

• ✅ No code changes (markdown only)
• ✅ No new dependencies
• ✅ No package.json / tsconfig changes
• ✅ Backward compatible with all ARC-1 tool versions that support the 12-tool surface (read-only path: SAPManage(probe), SAPRead, SAPSearch, SAPContext)
• ✅ Works with mcp-sap-docs MCP server (optional but recommended for Clean Core gap analysis)
• ✅ No new ACTION_POLICY entries needed
• ✅ No CI gate changes

Testing

This is a skill (markdown), no automated tests apply per arc-1 convention.

Smoke-testable manually by:

1. Connecting ARC-1 to a system with Z packages (read-only access sufficient)
2. Invoking the skill via Claude Code / Cursor / Copilot
3. Verifying scaffold output compiles: npx cds compile <target>/db/schema.cds --to edmx

Inspired by

• generate-rap-service-researched — research-first pattern
• sap-clean-core-atc — Clean Core classification structure
• migrate-custom-code — ATC-driven fix pattern

Open questions for the maintainer

1. Cluster placement: I created a new "BTP CAP Modernization (Preview)" section in skills/README.md. Open to feedback on:
   • Section name / position
   • Whether to mark "Preview" until all 6 sub-skills land
2. Naming convention: All new skills prefixed modernize-abap-*. Other options considered: port-abap-to-cap-*, btp-cap-from-abap-*. Prefer the current modernize-* family for symmetry with migrate-*.
3. Follow-up sub-skills strategy: Do you prefer (a) a single follow-up PR with all 4 sub-skills, (b) one PR per sub-skill in sequence, or (c) sub-skills can be PRs from community contributors after this MVP merges?
4. Smoke test on real system: I can run a real-world smoke test against a public-cloud trial tenant with a Z package. If you have a preferred test fixture / package I should target, happy to align.

Stats

• Files changed: 4 (3 new, 1 modified)
• Lines added: 1,247
• Lines removed: 0
• New code dependencies: 0
• New tools: 0
• New ACTION_POLICY entries: 0

---

Marked as Draft to invite review before finalization. Happy to address comments and revise.

🤖 Generated with Claude Code

[marianfoo]

Good idea! Will need further testing, but i think only skills update, from arc-1 side nothing to do

[marianfoo]

@Raistlin82
Please stop creating new PRs, use this PR only for adding new skills.
As I said, I would like to add only skills related to ARC-1, for CAP only skills maybe create a separate repo.
Also don't add those long comments to the PRs, I will not read them all, please write yourself, especially when you have a question or want something for me.