Sap Transport Gate — AI skill for Claude Code

sap-transport-gate

[English](README.md) | [中文](README.zh-CN.md)

[!IMPORTANT] **This repository has been archived and is no longer updated.** `sap-transport-gate` is now maintained exclusively in the [sap-engineering-skill](https://github.com/shrek-abaper/sap-engineering-skill) monorepo at [`skills/sap-transport-gate`](https://github.com/shrek-abaper/sap-engineering-skill/tree/main/skills/sap-transport-gate). Please go there for the latest version, issues, and improvements.

AI-assisted pre-release gate review for SAP Transport Requests.

An AI agent skill that performs structured, evidence-driven release readiness assessment for SAP Transport Requests. Produces an auditable `GO / CONDITIONAL_GO / NO_GO / NEED_MORE_EVIDENCE` decision and a formal Release Readiness Report.

---

What It Does

`sap-transport-gate` guides an AI agent through a structured review workflow:

1. Extract TR ID — Detect Transport Request ID from user input (pattern: uppercase letters + K + 6 digits, e.g., DEVK900123); ask if not found
2. Identify Review Mode — Offline Package, Offline Local, or Online Transport; verify SAP credentials for Online mode
3. Select Review Scope — User confirms: (A) Code Quality only or (B) Functional + Code Quality (spec required for B)
4. Evidence Intake — Inventory all provided materials; flag gaps
5. Evidence Level — HIGH / MEDIUM / LOW / UNKNOWN based on completeness
6. Multi-Dimension Review — 10 dimensions covering the full release risk surface
7. Finding Classification — Structured findings with severity, confidence, evidence, and recommendation
8. Release Decision — Evidence-based GO / CONDITIONAL_GO / NO_GO / NEED_MORE_EVIDENCE
9. Report Generation — Markdown Release Readiness Report + JSON summary

---

Core Principles

Principle	Rule
Evidence-first	AI never invents conclusions from insufficient evidence. Every finding must trace to real code, metadata, or material. Eve