abap-mock-usage

星标0

分支1

更新时间2026年6月16日 17:16

Differences and deterministic behavior of the mock ADT server used by adt-bench smoke runs.

安装

用 Codex 或 Claude 帮你安装复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。

在 Manus 中运行

来源

abapify

abapify/adt-bench

打开 GitHub 仓库查看创作者相关仓库

下载

在 Manus 中运行

abap-mock-usage

When the benchmark is running against the local mock ADT server (the default in v1), a few things differ from a real SAP system:

Authentication is always Basic (user demo, password demo). No X.509, no SNC, no OAuth.
CSRF tokens are issued by the mock on any GET with X-CSRF-Token: Fetch. POST/PUT currently do not enforce the token in v1.
Locks are per-object. A second LOCK from another session overwrites the handle in v1 (real SAP returns 200 with empty handle).
Activation always succeeds for any object that has a source. It does not actually run the ABAP compiler.
ABAP Unit is not implemented in v1 of the mock. Use the kind: "test" evidence only when the real backend is configured.
The __mock/* admin endpoints (/__mock/state, /__mock/reset, /__mock/stats) are not part of SAP ADT. Do not call them from production code; they exist only for the harness and the evaluator.
Statistics — /__mock/stats returns { total, by_endpoint } counters that the harness uses for the optional metrics.adt_http_calls field.
No transport checks — transportchecks is not implemented; the mock skips it and treats $TMP packages as local.

When a scenario is run in mock mode, the prompt may include a "Mock ADT server" section with the URL. Use that URL as the base for your MCP server's ABAP_URL (or equivalent).

同仓库更多 Skills

同仓库

agent-task

abapify/adt-bench

How to consume a GitHub issue body as an agent task prompt. The 9-section template, the lifecycle, the failure modes.

2026-06-160

principles

abapify/adt-bench

The 12 design principles this codebase is built on. Read first, before any task.

2026-06-160

sdd

abapify/adt-bench

Spec-driven development — the practice of writing the spec before the code. With the 9-section template and the three pnpm spec-check gates.

2026-06-160

abap-result-contract

abapify/adt-bench

How to return a strict scenario result JSON with evidence and no extra prose.

2026-06-160

abap-test-loop

abapify/adt-bench

Test-driven loop for ABAP — inspect, change, check syntax, run unit tests, diagnose, fix.

2026-06-160

abap-workflow

abapify/adt-bench

Standard ABAP object workflow (discover, create or edit, check, activate, validate).

2026-06-160

name	abap-mock-usage
description	Differences and deterministic behavior of the mock ADT server used by adt-bench smoke runs.

abap-mock-usage

When the benchmark is running against the local mock ADT server (the default in v1), a few things differ from a real SAP system:

Authentication is always Basic (user demo, password demo). No X.509, no SNC, no OAuth.
CSRF tokens are issued by the mock on any GET with X-CSRF-Token: Fetch. POST/PUT currently do not enforce the token in v1.
Locks are per-object. A second LOCK from another session overwrites the handle in v1 (real SAP returns 200 with empty handle).
Activation always succeeds for any object that has a source. It does not actually run the ABAP compiler.
ABAP Unit is not implemented in v1 of the mock. Use the kind: "test" evidence only when the real backend is configured.
The __mock/* admin endpoints (/__mock/state, /__mock/reset, /__mock/stats) are not part of SAP ADT. Do not call them from production code; they exist only for the harness and the evaluator.
Statistics — /__mock/stats returns { total, by_endpoint } counters that the harness uses for the optional metrics.adt_http_calls field.
No transport checks — transportchecks is not implemented; the mock skips it and treats $TMP packages as local.

When a scenario is run in mock mode, the prompt may include a "Mock ADT server" section with the URL. Use that URL as the base for your MCP server's ABAP_URL (or equivalent).