Vibing Steampunk (vsp)

AI-Agentic Development Unlocked for ABAP — ECC, S/4HANA, everywhere ADT is available.

ADT ↔ MCP Bridge: Gives Claude (and other AI assistants) full access to SAP ADT APIs. Read code, write code, debug, deploy, run tests — all through natural language (or DSL for automation).

See also: OData ↔ MCP Bridge for SAP data access.

Want to review or test? Start here: Reviewer Guide — 8 hands-on tasks, no SAP needed.

100 Stars!

Read the milestone article: Agentic ABAP at 100 Stars: The Numbers, The Community, and What's Cooking

What's New

v2.25.0 - CreatePackage Software Component Support

software_component Parameter: Create transportable packages with proper software component (e.g., HOME, ZLOCAL)
Viper Env Fix: Comma-separated env vars (SAP_ALLOWED_PACKAGES, SAP_ALLOWED_TRANSPORTS) now parse correctly
Closes #18: Namespace URL encoding verified working
⚠️ Requires: --allow-transportable-edits or --enable-transports flag for transportable packages

v2.24.0 - Transportable Edits Safety Feature

Safety by Default: Editing objects in transportable packages blocked unless explicitly enabled
--allow-transportable-edits: Opt-in flag to enable editing non-local objects
Transport Whitelisting: --allowed-transports "A4HK*,DEVK*" restricts allowed TRs
Package Whitelisting: --allowed-packages "Z*,$TMP" restricts editable packages
Clear Error Messages: Explains why edits are blocked and how to enable
Transport Tool Visibility: ListTransports/GetTransport visible when flag enabled
See Transportable Edits Report

v2.23.0 - GitExport to Disk & GetAbapHelp via WebSocket

GitExport saves ZIP to disk: No more base64 - files written directly to output_dir
GetAbapHelp via WebSocket: Real SAP documentation from system (uses ZADT_VSP if connected)
Lazy WebSocket Connection: WebSocket connects on-demand for Git/Help operations
Embedded ABAP Sync: Updated all ZADT_VSP classes from SAP system (2026-02-02)

v2.22.0 - SAP GUI Debugger Integration

SAP_TERMINAL_ID: Share breakpoints between vsp and SAP GUI sessions
MoveObject Tool: Move ABAP objects between packages (via ZADT_VSP WebSocket)
RunReport Refactor: Uses background jobs with spool output - more reliable execution
Cross-Tool Debugging: Set breakpoint in vsp, hit it in SAP GUI (or vice versa)
Configure: --terminal-id <id> or SAP_TERMINAL_ID=<id> (get ID from SAP GUI registry/file)

v2.21.0 - Method-Level Source Operations

GetSource with method: Returns only the METHOD...ENDMETHOD block (~50 lines vs 1000+)
EditSource with method: Constrains find/replace to specific method - no accidental edits elsewhere
WriteSource with method: Replace only one method implementation (method must exist)
95% Token Reduction: Work with individual methods instead of entire classes
AI-Optimized: Faster responses, more precise edits, safer refactoring
See Method-Level Operations Report

v2.20.0 - CLI Mode & Multi-System Management

CLI Subcommands: vsp search, vsp source, vsp export, vsp systems - direct ABAP operations without MCP
System Profiles: .vsp.json config file for managing multiple SAP systems
Config Conversion: vsp config mcp-to-vsp / vsp-to-mcp - convert between config formats
Cookie Auth in Profiles: Support cookie_file / cookie_string for SSO environments
Password Sync: Passwords imported from .mcp.json env blocks automatically
GitExport FULL Mode: Multi-package exports with proper folder hierarchy (src/$package/)
abapGit Compatible: Exports now include .abapgit.xml with FULL folder logic

v2.19.0 - Async Execution & Developer Productivity

RunReportAsync: Execute reports in background goroutine - no more timeouts!
GetAsyncResult: Poll or wait for async task completion (up to 60s)
CompareSource: Unified diff between any two ABAP objects (LCS algorithm)
CloneObject: Copy PROG/CLAS/INTF to new name with auto-replace
GetClassInfo: Quick class metadata via CAI (methods, attrs, interfaces)
CreateTable: Create DDIC tables from simple JSON definition
GetSystemInfo: Fixed to use SQL (CVERS/T000) - works across all SAP versions
54 Focused / 99 Expert Tools - see Release Notes

v2.18.1 - Interactive CLI Debugger

vsp debug Command: Standalone interactive ABAP debugger
REPL Interface: s=step, n=next, o=out, c=continue, r=stack, v=vars
WebSocket Breakpoints: Set/delete/list via ZADT_VSP
Attach Mode: Wait for any debuggee or filter by user
Graceful Handling: Ctrl+C cleanup, session state management
Phase 1 of DAP implementation - see DAP Plan

v2.18.0 - Report Execution Tools

RunReport: Execute selection-screen reports with params/variants, capture ALV output
GetVariants: List available variants for a report
GetTextElements: Read selection texts and text symbols
SetTextElements: Update selection texts and text symbols programmatically
Tool Group "R": Disable report tools with --disabled-groups R
ZADT_VSP v2.3.0: New "report" domain for WebSocket-based report execution
See Report Design

v2.17.0 - Install Tools & One-Command Deployment

InstallZADTVSP: Deploy WebSocket handler to SAP in one command
InstallAbapGit: Deploy abapGit standalone or dev edition
ListDependencies: Show available installable packages
Upsert Logic: Proper CREATE vs UPDATE detection for all objects
Embedded ABAP source in binary - no external files needed

v2.16.0 - abapGit WebSocket Integration

GitTypes Tool: Query 158 supported abapGit object types
GitExport Tool: Export packages/objects as abapGit-compatible ZIP (base64)
WebSocket Transport: Via ZADT_VSP handler (domain: "git")
abapGit Serialization: Uses native ZCL_ABAPGIT_OBJECTS=>serialize() for full compatibility
Tool Group "G": Disable Git tools with --disabled-groups G
Requires: abapGit installed on SAP system
See abapGit Integration Report

v2.15.0 - Phase 5 TAS-Style Debugging Complete

Variable History Recording: Track all variable changes during debug sessions
Extended Breakpoint Types: Statement, exception, and watchpoint scripting
Force Replay / State Injection: THE KILLER FEATURE - inject captured state into live debug sessions
WebSocket Debugging: Full TPDAPI integration via ZADT_VSP WebSocket handler
AMDP Debugging: Experimental support (session works, breakpoint triggering under investigation)
See Observations Since v2.12.5 for complete changelog

v2.14.0 - Lua Scripting Integration (Phase 5)

vsp lua Command: Interactive REPL and script execution
40+ Lua Bindings: All MCP tools accessible from Lua (searchObject, getSource, setBreakpoint, etc.)
Debug Session Management: listen, attach, stepOver, stepInto, stepReturn, continue, getStack
Checkpoint System: saveCheckpoint, getCheckpoint, listCheckpoints, injectCheckpoint (Force Replay foundation)
Example Scripts: search-and-grep, call-graph-analysis, debug-session, analyze-dumps
See TAS & RCA Vision and Roadmap for Phase 5-8 plans

v2.13.0 - Call Graph & RCA Tools

GetCallersOf/GetCalleesOf: Navigate call graphs up (who calls) and down (what's called)
TraceExecution: Composite RCA tool - static graph + trace + comparison in one call
CompareCallGraphs: Find untested paths (static only) and dynamic calls (actual only)
AnalyzeCallGraph: Statistics on nodes, edges, depth, types
WebSocket Debugging: Full TPDAPI integration for statement/exception breakpoints

v2.12.5 - EditSource Line Ending Fix

CRLF→LF Normalization: EditSource now works reliably across platforms
SAP ADT returns \r\n, AI sends \n - now automatically normalized
No more "old_string not found" errors due to line ending mismatches

v2.12.4 - Feature Detection & Safety Network

GetFeatures Tool: Probe SAP system for available optional capabilities
Feature Flags: --feature-abapgit, --feature-rap, --feature-amdp, --feature-ui5, --feature-transport
Feature Modes: auto (probe system), on (force enable), off (force disable)
SRVB WriteSource: Create Service Bindings via unified WriteSource tool
BDEF Improvements: Fixed behavior definition creation flow

v2.12.0 - abapGit-Compatible Format & Batch Operations

Class Includes: Import/export .clas.testclasses.abap, .clas.locals_def.abap, .clas.locals_imp.abap
Batch Import DSL: dsl.Import(client).FromDirectory("./src/").RAPOrder().Execute(ctx)
Batch Export DSL: dsl.Export(client).Classes("ZCL_*").ToDirectory("./backup/").Execute(ctx)
Pipeline Builder: dsl.RAPPipeline(client, "./src/", "$PKG", "ZSRV_BINDING") - full RAP deployment
Priority Control: DDLSFirst(), RAPOrder(), CustomOrder() for dependency handling

v2.11.0 - Transport Management & Safety Controls

5 Transport Tools: ListTransports, GetTransport, CreateTransport, ReleaseTransport, DeleteTransport
Safety Controls: --enable-transports, --allowed-transports "A4HK*", --allowed-packages "Z*"
Tool Group "C": Disable all CTS tools with --disabled-groups C
Enterprise-grade transport governance for AI assistants
See also v2.24.0 for --allow-transportable-edits safety feature

v2.10.0 - UI5/BSP Management & Tool Groups

7 UI5/BSP Tools: List apps, read files, search content, view manifests
AMDP Debugger: 5 tools for HANA stored procedure debugging
Tool Groups: Selectively disable features (--disabled-groups 5THD)
94 Total Tools: 46 focused mode, 94 expert mode

v2.8.0 - Full Debug Session Support

DebuggerAttach/Detach - Attach to caught debuggees, release sessions
DebuggerGetStack - View call stack with program/line info
DebuggerGetVariables - Inspect variable values during debugging
DebuggerStep - Step into/over/return/continue through code
Complete AI-powered debugging: breakpoint → listen → attach → inspect → step

v2.7.0 - External Debugger & Listener

Set external breakpoints (line, exception, statement, message)
Long-polling debug listener for autonomous debugging
Foundation for AI-powered debugger scripting

v2.6.0 - RAP OData E2E Support

Create CDS views, Service Definitions, and Service Bindings
Publish OData V2/V4 services directly from AI assistant
Full RAP development lifecycle via MCP tools

Single binary with 81 focused tools (default) or 122 expert tools for AI-assisted ABAP development.

Key Features

Feature	Description
AI Debugger	Breakpoints, listener, attach, step, inspect stack & variables
RAP OData E2E	Create CDS views, Service Definitions, Bindings → Publish OData services
Focused Mode	37 curated tools optimized for AI assistants (50% token reduction)
AI-Powered RCA	Root cause analysis with dumps, traces, profiler + code intelligence
DSL & Workflows	Fluent Go API + YAML automation for CI/CD pipelines
ExecuteABAP	Run arbitrary ABAP code via unit test wrapper
Code Analysis	Call graphs, object structure, find definition/references
System Introspection	System info, installed components, CDS dependencies
Diagnostics	Short dumps (RABAX), ABAP profiler (ATRA), SQL traces (ST05)
File Deployment	Bypass token limits - deploy large files directly from filesystem
Surgical Edits	`EditSource` tool matches Claude's Edit pattern for precise changes

Quick Start

#Download binary from releases
curl -LO https://github.com/oisee/vibing-steampunk/releases/latest/download/vsp-linux-amd64
chmod +x vsp-linux-amd64

#Or build from source
git clone https://github.com/oisee/vibing-steampunk.git && cd vibing-steampunk
make build

Windows 11 with VS Code + Claude Code extension:

1. Get the latest vsp release:

https://github.com/oisee/vibing-steampunk/releases.

If you have trouble downloading executable files in your browser, use curl -o url or wget to download the file. Name the file vsp.exe.

Put the file in a local folder and open the folder in VS Code.

Add the vsp folder to your PATH environment variable for your user. Either through command line or Windows Registry Editor regedit. Add the vsp folder to KEY_CURRENT_USER\Environment\Path.

Restart your VS Code to recognize the updated PATH before progressing to the next steps.

2. Initialize the config files:

Open a terminal in VS Code, then run ./vsp config init to create config template files:

.env.example
.vsp.json.example
.mcp.json.example

3. Adjust your config files:

Make sure you delete the comment lines. Refer to the example files in this README.

4. Set up authentication

For basic auth: Set up a password for your user to allow for basic authentication. Go to SU01 > Logon Data, generate an initial password. Then log in again (without SNC/SSO in SAPGUI) and change the initial password. You are now set up for basic authentication via config file in vsp. Set your environment variables SAP_USER and SAP_PASSWORD accordingly.

You now need to obtain the SAP hostname for your SAP_URL environment variable: Log in to any web-based application (e.g. Fiori Launchpad) and obtain the URL from your browser. Attention: SAP_URL is not not your message/group server from SAP Logon!

Alternatively use cookie authentication: If you cannot set a password for your user, you may still use cookie authentication to access your SAP system from vsp.

Extract cookies manually and save them in cookies.txt in your vsp folder. Use cookies SAP_SESSIONID_SYS_CLI and sap-usercontext on your previously determined URL (caution: use https:// prefix for secure connections). Refer to below guide on how to manually extract cookies from your browser.

Template cookie file:

# Netscape HTTP Cookie File
# https://curl.haxx.se/rfc/cookie_spec.html

https://your.domain.com	FALSE	/	TRUE	0	SAP_SESSIONID_SYS_CLI  YOUR_CONTENT
https://your.domain.com	FALSE	/	TRUE	0	sap-usercontext        YOUR_CONTENT

Replace the hostname, SYS with your system ID (e.g. DS1) and CLI with your client number (e.g. 100).

For BTP/Cloud based systems: Use cookies __VCAP_ID__ and JSESSIONID on your domain https://xyz.ondemand.com. This also works for BTP trial accounts. Also refer to this article.

Obtaining cookies from your browser session:

The easiest way to do so is to use Edge as it allows you to display cookie contents from its settings page. From there you can copy & paste them into the newly created cookies.txt file.

Open any transaction in WebDynpro or Fiori Launchpad. For older environments like ECC it should work with BRF+ transactions that open in a browser. Login with your credentials. Once logged in it’s a matter of extracting the created session cookies.

In Edge, go to Settings > Privacy, search and services > Cookies > See all cookies and site data. Search for your top-level domain. There should be two cookies for your system as described above (SAP_SESSIONID and sap-usercontext or VCAP_ID and JSESSIONID if your system is cloud-based). Copy the content values for each cookie to your local file and save.

The created cookies are session cookies. They will eventually expire after a timeout and the values in cookies.txt need to be updated. Usually Claude will tell you if this is the case.

5. Test the connection:

Use the terminal with this command: ./vsp -s dev search "zcl_*" --type CLAS --max 50.

You will get prompted with a list of found objects if the connection could be established.

CLI Coding Agents

VSP works with 8 CLI coding agents — not just Claude! Full setup guides with config templates:

Agent	LLM	Free?	Config
Gemini CLI	Gemini 2.5 Pro/Flash	Yes (1000 req/day)	`.gemini/settings.json`
Claude Code	Claude Opus/Sonnet 4.6	No ($20+/mo)	`.mcp.json`
GitHub Copilot	Claude, GPT-5, Gemini	No ($10+/mo)	`.copilot/mcp-config.json`
OpenAI Codex	GPT-5-Codex, GPT-4.1	No ($20+/mo)	`.mcp.json`
Qwen Code	Qwen3-Coder	Yes (1000 req/day)	`.qwen/settings.json`
OpenCode	75+ models (BYOK)	Yes (own key)	`opencode.json`
Goose	75+ providers (BYOK)	Yes (own key)	`~/.config/goose/config.yaml`
Mistral Vibe	Devstral 2, local models	Yes (Ollama)	`.vibe/config.toml`

Full setup guide with config examples | Русский | Українська | Español

CLI Mode

vsp works in two modes:

MCP Server Mode (default) - Exposes tools via Model Context Protocol for Claude
CLI Mode - Direct command-line operations without MCP

CLI Commands

# Search for ABAP objects
vsp -s a4h search "ZCL_*"
vsp -s dev search "Z*ORDER*" --type CLAS --max 50

# Get source code
vsp -s a4h source CLAS ZCL_MY_CLASS
vsp -s dev source PROG ZTEST_PROGRAM

# Export packages to ZIP (abapGit format)
vsp -s a4h export '$ZORK' '$ZLLM' -o packages.zip
vsp -s dev export '$TMP' --subpackages

# List configured systems
vsp systems

# Configuration management
vsp config init          # Create example configs
vsp config show          # Show effective configuration
vsp config mcp-to-vsp    # Import from .mcp.json to .vsp.json
vsp config vsp-to-mcp    # Export from .vsp.json to .mcp.json

System Profiles (`.vsp.json`)

Configure multiple SAP systems in .vsp.json:

{
  "default": "dev",
  "systems": {
    "dev": {
      "url": "http://dev.example.com:50000",
      "user": "DEVELOPER",
      "client": "001"
    },
    "a4h": {
      "url": "http://a4h.local:50000",
      "user": "ADMIN",
      "client": "001",
      "insecure": true
    },
    "prod": {
      "url": "https://prod.example.com:44300",
      "user": "READONLY",
      "client": "100",
      "read_only": true,
      "cookie_file": "/path/to/cookies.txt"
    }
  }
}

Password Resolution:

Set via environment variable: VSP_<SYSTEM>_PASSWORD (e.g., VSP_DEV_PASSWORD)
Or use cookie authentication: cookie_file or cookie_string

Config Locations (searched in order):

.vsp.json (current directory)
.vsp/systems.json
~/.vsp.json
~/.vsp/systems.json

MCP Server Configuration

CLI Flags

vsp --url https://host:44300 --user admin --password secret
vsp --url https://host:44300 --cookie-file cookies.txt
vsp --mode expert  # Enable all 122 tools

Environment Variables

export SAP_URL=https://host:44300
export SAP_USER=developer
export SAP_PASSWORD=secret
export SAP_CLIENT=001

.env File

# .env (auto-loaded from current directory)
SAP_URL=https://host:44300
SAP_USER=developer
SAP_PASSWORD=secret

Flag	Env Variable	Description
`--url`	`SAP_URL`	SAP system URL
`--user`	`SAP_USER`	Username
`--password`	`SAP_PASSWORD`	Password
`--client`	`SAP_CLIENT`	Client (default: 001)
`--mode`	`SAP_MODE`	`focused` (default) or `expert`
`--cookie-file`	`SAP_COOKIE_FILE`	Netscape cookie file
`--insecure`	`SAP_INSECURE`	Skip TLS verification
`--terminal-id`	`SAP_TERMINAL_ID`	SAP GUI terminal ID for cross-tool debugging
`--allow-transportable-edits`	`SAP_ALLOW_TRANSPORTABLE_EDITS`	Enable editing transportable objects
`--allowed-transports`	`SAP_ALLOWED_TRANSPORTS`	Whitelist transports (wildcards: `A4HK*`)
`--allowed-packages`	`SAP_ALLOWED_PACKAGES`	Whitelist packages (wildcards: `Z*,$TMP`)

Usage with Claude

Claude Desktop

Add to ~/.config/claude/claude_desktop_config.json:

{
  "mcpServers": {
    "abap-adt": {
      "command": "/path/to/vsp",
      "env": {
        "SAP_URL": "https://your-sap-host:44300",
        "SAP_USER": "your-username",
        "SAP_PASSWORD": "your-password"
      }
    }
  }
}

Claude Code

Add .mcp.json to your project:

{
  "mcpServers": {
    "abap-adt": {
      "command": "/path/to/vsp",
      "env": {
        "SAP_URL": "https://your-sap-host:44300",
        "SAP_USER": "your-username",
        "SAP_PASSWORD": "your-password"
      }
    }
  }
}

Transportable Packages Configuration

To work with transportable packages (non-$ prefixed), you must explicitly enable transport support:

{
  "mcpServers": {
    "abap-adt": {
      "command": "/path/to/vsp",
      "env": {
        "SAP_URL": "https://your-sap-host:44300",
        "SAP_USER": "your-username",
        "SAP_PASSWORD": "your-password",
        "SAP_CLIENT": "001",
        "SAP_ALLOW_TRANSPORTABLE_EDITS": "true",
        "SAP_ALLOWED_TRANSPORTS": "DEVK*,A4HK*",
        "SAP_ALLOWED_PACKAGES": "ZPROD,$TMP,$*,Z*"
      }
    }
  }
}

Env Variable	Purpose
`SAP_ALLOW_TRANSPORTABLE_EDITS`	Enable editing objects in transportable packages
`SAP_ENABLE_TRANSPORTS`	Enable full transport management (create, release)
`SAP_ALLOWED_TRANSPORTS`	Whitelist transport patterns (wildcards supported)
`SAP_ALLOWED_PACKAGES`	Whitelist package patterns (wildcards supported)

CreatePackage with software component:

CreatePackage(
  name="ZPROD_005",
  description="Sub-package",
  parent="ZPROD",
  transport="DEVK900123",
  software_component="HOME"
)

Without these flags, operations on transportable packages will be blocked by the safety system.

Focused vs Expert Mode

Aspect	Focused (Default)	Expert
Tools	52 essential	99 complete
Token overhead	~2,800	~8,000
Use case	Daily development	Edge cases, debugging
Unified tools	GetSource, WriteSource	+ granular Get/Write

Focused mode consolidates 11 read/write tools into 2 unified tools, reducing cognitive load and token usage by 58%.

Enable expert mode: --mode=expert or SAP_MODE=expert

DSL & Automation

YAML Workflows

# ci-pipeline.yaml
name: CI Pipeline
vars:
  package: "$TMP"
steps:
  - action: search
    query: "ZCL_*"
    types: [class]
    package: "{{ .package }}"
    save_as: classes

  - action: test
    objects: "{{ .classes }}"
    parallel: 4

  - action: fail_if
    condition: tests_failed
    message: "Unit tests failed"

vsp workflow run ci-pipeline.yaml --var package='$ZRAY'

Go Library

// Fluent search
objects, _ := dsl.Search(client).
    Query("ZCL_*").Classes().InPackage("$TMP").Execute(ctx)

// Test orchestration
summary, _ := dsl.Test(client).
    Objects(objects...).Parallel(4).Run(ctx)

// Batch import from directory (abapGit-compatible)
result, _ := dsl.Import(client).
    FromDirectory("./src/").
    ToPackage("$ZRAY").
    RAPOrder().  // DDLS → BDEF → Classes → SRVD
    Execute(ctx)

// Export classes with all includes
result, _ := dsl.Export(client).
    Classes("ZCL_TRAVEL", "ZCL_BOOKING").
    ToDirectory("./backup/").
    Execute(ctx)

// RAP deployment pipeline
pipeline := dsl.RAPPipeline(client, "./src/", "$ZRAY", "ZTRAVEL_SB")

See docs/DSL.md for complete documentation.

RAP OData Service Creation

VSP supports full RAP OData E2E development since v2.6.0. Create complete OData services via AI assistant:

Step-by-Step Workflow

1. Create CDS View (DDLS)

WriteSource(
  object_type="DDLS",
  name="ZTRAVEL",
  package="$TMP",
  description="Travel Entity",
  source=`
@EndUserText.label: 'Travel'
@AccessControl.authorizationCheck: #NOT_REQUIRED
define root view entity ZTRAVEL as select from ztravel_tab {
  key travel_id as TravelId,
  description as Description,
  start_date as StartDate,
  end_date as EndDate,
  status as Status
}
`
)

2. Create Behavior Definition (BDEF)

WriteSource(
  object_type="BDEF",
  name="ZTRAVEL",
  package="$TMP",
  description="Travel Behavior",
  source=`
managed implementation in class ZBP_TRAVEL unique;
strict ( 2 );

define behavior for ZTRAVEL alias Travel
persistent table ztravel_tab
lock master
authorization master ( instance )
{
  field ( readonly ) TravelId;
  field ( mandatory ) Description;

  create;
  update;
  delete;

  mapping for ztravel_tab {
    TravelId = travel_id;
    Description = description;
    StartDate = start_date;
    EndDate = end_date;
    Status = status;
  }
}
`
)

3. Create Service Definition (SRVD)

WriteSource(
  object_type="SRVD",
  name="ZTRAVEL_SD",
  package="$TMP",
  description="Travel Service Definition",
  source=`
@EndUserText.label: 'Travel Service'
define service ZTRAVEL_SD {
  expose ZTRAVEL;
}
`
)

4. Create Service Binding (SRVB)

WriteSource(
  object_type="SRVB",
  name="ZTRAVEL_SB",
  package="$TMP",
  description="Travel OData V4 Binding",
  service_definition="ZTRAVEL_SD",
  binding_version="V4"
)

Binding Options

Parameter	Values	Description
`binding_version`	`V2`, `V4`	OData protocol version
`binding_category`	`0`, `1`	`0`=Web API, `1`=UI

For Transportable Packages

Add transport parameter to all WriteSource calls:

WriteSource(
  object_type="DDLS",
  name="ZTRAVEL",
  package="ZPROD",
  transport="DEVK900123",
  ...
)

ExecuteABAP

Run arbitrary ABAP code via unit test wrapper:

ExecuteABAP:
  code: |
    DATA(lv_msg) = |Hello from SAP at { sy-datum }|.
    lv_result = lv_msg.

Risk levels: harmless (read-only), dangerous (write), critical (full access)

See ExecuteABAP Report for details.

AI-Powered Root Cause Analysis

vsp enables AI assistants to investigate production issues autonomously:

User: "Investigate the ZERODIVIDE crash in production"

AI Workflow:
  1. GetDumps      → Find recent crashes by exception type
  2. GetDump       → Analyze stack trace and variable values
  3. GetSource     → Read code at crash location
  4. GetCallGraph  → Trace call hierarchy
  5. GrepPackages  → Find similar patterns
  6. Analysis      → Identify root cause
  7. Propose Fix   → Generate solution + test case

Example Output:

"The crash occurs in ZCL_PRICING=>CALCULATE_RATIO when LV_TOTAL=0. This happens for archived orders with no line items. Here's the fix..."

See AI-Powered RCA Workflows for the complete vision.

Tools Reference

52 Focused Mode Tools:

Search: SearchObject, GrepObjects, GrepPackages
Read: GetSource, GetTable, GetTableContents, RunQuery, GetPackage, GetFunctionGroup, GetCDSDependencies
Debugger: DebuggerListen, DebuggerAttach, DebuggerDetach, DebuggerStep, DebuggerGetStack, DebuggerGetVariables
- Note: Breakpoints now managed via WebSocket (ZADT_VSP)
Write: WriteSource, EditSource, ImportFromFile, ExportToFile, MoveObject
Dev: SyntaxCheck, RunUnitTests, RunATCCheck, LockObject, UnlockObject
Intelligence: FindDefinition, FindReferences
System: GetSystemInfo, GetInstalledComponents, GetCallGraph, GetObjectStructure, GetFeatures
Diagnostics: GetDumps, GetDump, ListTraces, GetTrace, GetSQLTraceState, ListSQLTraces
Git: GitTypes, GitExport (requires abapGit on SAP)
Reports: RunReport, GetVariants, GetTextElements, SetTextElements
Install: InstallZADTVSP, InstallAbapGit, ListDependencies

See README_TOOLS.md for complete tool documentation (122 tools).

Capability Matrix

Capability	ADT (Eclipse)	abap-adt-api (TS)	vsp
Programs, Classes, Interfaces	Y	Y	Y
Functions, Function Groups	Y	Y	Y
Tables, Structures	Y	Y	Y
CDS Views	Y	Y	Y
Syntax Check, Activation	Y	Y	Y
Unit Tests	Y	Y	Y
CRUD Operations	Y	Y	Y
Find Definition/References	Y	Y	Y
Code Completion	Y	Y	Y
ATC Checks	Y	Y	Y
Call Graph	Y	Y	Y
System Info	Y	Y	Y
Surgical Edit (Edit pattern)	-	-	Y
File-based Deploy	-	-	Y
ExecuteABAP	-	-	Y
RAP OData (DDLS/SRVD/SRVB)	Y	-	Y
OData Service Publish	Y	-	Y
abapGit Export	Y	-	Y (WebSocket)
Debugging	Y	Y	N

Credits

Project	Author	Contribution
abap-adt-api	Marcello Urbani	TypeScript ADT library, definitive API reference
mcp-abap-adt	Mario Andreschak	First MCP server for ABAP ADT

vsp is a Go rewrite with:

Single binary, zero dependencies
62 tools (vs 13 original)
~50x faster startup

Optional: WebSocket Handler (ZADT_VSP)

vsp can optionally deploy a WebSocket handler to SAP for enhanced functionality like RFC calls:

# 1. Create package
vsp CreatePackage --name '$ZADT_VSP' --description 'VSP WebSocket Handler'

# 2. Deploy objects (embedded in binary)
vsp WriteSource --object_type INTF --name ZIF_VSP_SERVICE --package '$ZADT_VSP' \
    --source "$(cat embedded/abap/zif_vsp_service.intf.abap)"
vsp WriteSource --object_type CLAS --name ZCL_VSP_RFC_SERVICE --package '$ZADT_VSP' \
    --source "$(cat embedded/abap/zcl_vsp_rfc_service.clas.abap)"
vsp WriteSource --object_type CLAS --name ZCL_VSP_APC_HANDLER --package '$ZADT_VSP' \
    --source "$(cat embedded/abap/zcl_vsp_apc_handler.clas.abap)"

# 3. Manually create APC app in SAPC + activate in SICF
#    See embedded/abap/README.md for details

After deployment, connect via WebSocket to call RFCs:

{"id":"1","domain":"rfc","action":"call","params":{"function":"BAPI_USER_GET_DETAIL","USERNAME":"TESTUSER"}}

See WebSocket Handler Report for complete documentation.

Documentation

Document	Description
docs/architecture.md	Architecture diagrams (Mermaid)
README_TOOLS.md	Complete tool reference (94 tools)
MCP_USAGE.md	AI agent usage guide
docs/DSL.md	DSL & workflow documentation
ARCHITECTURE.md	Technical architecture (detailed)
CLAUDE.md	AI development guidelines
embedded/abap/README.md	WebSocket handler deployment
docs/cli-agents/	CLI coding agents setup guide (8 agents, 4 languages)
Roadmap: Quick/Mid/Far Wins	Prioritized feature backlog
Observations Since v2.12.5	Recent changes & research summary

SQL Query Notes

Uses ABAP SQL syntax, not standard SQL:

Feature	Status
`ORDER BY col ASCENDING`	Works
`ORDER BY col DESCENDING`	Works
`ORDER BY col ASC/DESC`	FAILS - use ASCENDING/DESCENDING
`LIMIT n`	FAILS - use `max_rows` parameter

Development

# Build
make build          # Current platform
make build-all      # All 9 platforms

# Test
go test ./...                              # Unit tests (249)
go test -tags=integration -v ./pkg/adt/    # Integration tests (21+)

Architecture

vibing-steampunk/
├── cmd/vsp/main.go           # CLI (cobra/viper)
├── pkg/adt/
│   ├── client.go             # ADT client + read ops
│   ├── crud.go               # CRUD operations
│   ├── devtools.go           # Syntax check, activate, tests
│   ├── codeintel.go          # Definition, refs, completion
│   ├── workflows.go          # High-level workflows
│   └── http.go               # HTTP transport (CSRF, auth)
├── internal/mcp/server.go    # MCP tool handlers (62 tools)
└── pkg/dsl/                  # DSL & workflow engine

Project Status

Metric	Value
Tools	122 (81 focused, 122 expert)
Unit Tests	270+
Platforms	9 (Linux, macOS, Windows × amd64/arm64/386)

Roadmap

Completed (v2.15.0)

Parked (Needs Further Work)

AMDP Debugger - Experimental: Session works, breakpoint triggering under investigation (Report)
UI5/BSP Write - ADT filestore is read-only, needs custom plugin via /UI5/CL_REPOSITORY_LOAD
abapGit Export - WebSocket integration complete (v2.16.0) - GitTypes, GitExport tools (Report)
abapGit Import - Requires ZCL_ABAPGIT_OBJECTS=>deserialize with virtual repository

Planned

API Release State (ARS) - Contract stability checks
Message Server Logs
Background Job Management

Future Considerations

AMDP Session Persistence (enable full HANA debugging)
Graph Traversal & Analysis (code dependency graphs)
Test Intelligence (smart test execution based on changes)
Standard API Surface Scraper

Research Reports:

AMDP Session Architecture - Session binding analysis & solutions
Native ADT Features - Comprehensive ADT capability analysis
ADT Debugger API - External debugging REST API
AI-Powered RCA - Vision for AI-assisted debugging

Lua Scripting (New in v2.14)

Automate debugging workflows with Lua scripts:

# Interactive REPL
vsp lua

# Run a script
vsp lua examples/scripts/debug-session.lua

# Execute inline
vsp lua -e 'print(json.encode(searchObject("ZCL_*", 10)))'

Example: Set breakpoint and debug

-- Set breakpoint
local bpId = setBreakpoint("ZTEST_PROGRAM", 42)
print("Breakpoint: " .. bpId)

-- Wait for debuggee
local event = listen(60)
if event then
    attach(event.id)
    print("Stack:")
    for i, frame in ipairs(getStack()) do
        print("  " .. frame.program .. ":" .. frame.line)
    end
    stepOver()
    detach()
end

Available Functions:

Search: searchObject, grepObjects
Source: getSource, writeSource, editSource
Debug: setBreakpoint, listen, attach, detach, stepOver, stepInto, stepReturn, continue_, getStack, getVariables
Checkpoints: saveCheckpoint, getCheckpoint, listCheckpoints, injectCheckpoint
Diagnostics: getDumps, getDump, runUnitTests, syntaxCheck
Call Graph: getCallGraph, getCallersOf, getCalleesOf
Utilities: print, sleep, json.encode, json.decode

See examples/scripts/ for more examples.

RCA, Replay & Test Extraction

The Vision: AI-Powered Debugging Pipeline

┌─────────────────────────────────────────────────────────────────────────────┐
│  1. SET BREAKPOINT    →  2. RUN PROGRAM    →  3. CAPTURE CONTEXT           │
│     setBreakpoint()       (trigger via         saveCheckpoint()             │
│     on FM/method          unit test/RFC)       for each hit                 │
├─────────────────────────────────────────────────────────────────────────────┤
│  4. EXTRACT TEST CASES  →  5. AI NORMALIZE  →  6. GENERATE UNIT TESTS      │
│     inputs + outputs       deduplicate,         ABAP Unit classes           │
│     from checkpoints       explain patterns     with mocks                  │
└─────────────────────────────────────────────────────────────────────────────┘

Example: Capture FM Execution for Test Generation

-- Step 1: Set breakpoint on function module entry
local bpId = setBreakpoint("SAPL<FGROUP>", 10)  -- Entry point

-- Step 2: Prepare to capture multiple executions
local captures = {}

-- Step 3: Loop to capture test cases
for i = 1, 10 do
    local event = listen(120)  -- Wait for debuggee
    if not event then break end

    attach(event.id)

    -- Capture input parameters at entry
    local vars = getVariables()
    local testCase = {
        id = i,
        inputs = extractInputs(vars),  -- IV_*, IT_*, IS_*
        timestamp = os.time()
    }

    -- Step to end to capture outputs
    continue_()
    local event2 = listen(5)
    if event2 then
        attach(event2.id)
        testCase.outputs = extractOutputs(getVariables())  -- EV_*, ET_*, ES_*, RETURN
    end

    -- Save checkpoint for replay
    saveCheckpoint("testcase_" .. i, testCase)
    table.insert(captures, testCase)

    detach()
end

-- Step 4: Export for AI processing
print(json.encode(captures))

AI Processing Pipeline

After capturing test cases, AI can:

Normalize & Deduplicate - Group similar inputs, identify unique scenarios
Explain Patterns - "TestCase 3 tests error path when IV_AMOUNT < 0"
Generate Unit Tests - Create ABAP Unit test class with proper mocks

User: "Analyze captured test cases and generate unit tests"

AI Workflow:
  1. Load checkpoints     → listCheckpoints("testcase_*")
  2. Analyze patterns     → Cluster by input signatures
  3. Identify edge cases  → Empty tables, zero values, error conditions
  4. Generate mock specs  → Which FMs/DB tables need mocking
  5. Create ABAP Unit     → ZCL_TEST_<FM> with test methods
  6. Deploy tests         → WriteSource to SAP system

What Works Today (v2.14)

Feature	Status	Command/Function
Set breakpoints	✅	`setBreakpoint(program, line)`
Listen for debuggee	✅	`listen(timeout)`
Attach/detach	✅	`attach(id)`, `detach()`
Step execution	✅	`stepOver()`, `stepInto()`, `continue_()`
Get variables	✅	`getVariables()`
Get stack trace	✅	`getStack()`
Save checkpoints	✅	`saveCheckpoint(name, data)`
Load checkpoints	✅	`getCheckpoint(name)`
Call graph analysis	✅	`getCallersOf()`, `getCalleesOf()`
Short dump analysis	✅	`getDumps()`, `getDump(id)`

Coming in Future Phases

Feature	Phase	Description
Variable history recording	5.2	✅ Track all variable changes during execution
Force Replay (state injection)	5.5	✅ Inject saved state into live debug session
Test case extraction	6.1	Automated input/output extraction from recordings
ABAP test generator	6.3	Generate ABAP Unit classes from test cases
Mock framework	6.4	ZCL_VSP_MOCK for DB/RFC mocking
Isolated playground	7.1	Fast test execution with mocked dependencies
Time-travel debugging	8.1	Navigate backwards through execution

Vision & Roadmap

Where we're going: TAS-style debugging, time-travel, AI-powered RCA

Phase	Target	Features
5	Q1 2026	Lua scripting ✅, variable history, checkpoints, Force Replay
6	Q2 2026	Test case extraction, ABAP test generator, mock framework
7	Q3 2026	Isolated playground with mocks, patch & re-run
8	Q4 2026	Time-travel debugging, temporal queries
9+	2027	AI-suggested breakpoints, multi-agent debugging, self-healing

Read more:

VISION.md - The dream: AI as a senior developer
ROADMAP.md - Detailed implementation plan
TAS & Scripting Report - Full technical design
Test Extraction Report - Playground architecture

License

MIT

Contributing

Contributions welcome! See ARCHITECTURE.md and CLAUDE.md for guidelines.

Name		Name	Last commit message	Last commit date
Latest commit History 205 Commits
.claude/commands		.claude/commands
abap/src/zadt_vsp		abap/src/zadt_vsp
articles		articles
cmd/vsp		cmd/vsp
docs		docs
embedded		embedded
examples		examples
internal/mcp		internal/mcp
media		media
pkg		pkg
reports		reports
scripts		scripts
src		src
.abapgit.xml		.abapgit.xml
.gitignore		.gitignore
.golangci.yml		.golangci.yml
ARCHITECTURE.md		ARCHITECTURE.md
CLAUDE.md		CLAUDE.md
MCP_USAGE.md		MCP_USAGE.md
Makefile		Makefile
README.md		README.md
README_TOOLS.md		README_TOOLS.md
ROADMAP.md		ROADMAP.md
VISION.md		VISION.md
amdp-breakpoint-test-results-20251206-084536.log		amdp-breakpoint-test-results-20251206-084536.log
go.mod		go.mod
go.sum		go.sum

Document	Description
VISION.md	The dream: AI as a senior developer
ROADMAP.md	Detailed implementation timeline
TAS & Scripting	Technical design for TAS-style debugging
Test Extraction	Playground and mock architecture
Force Replay	State injection design
Implications Analysis	Paradigm shift: archaeology → observation
AI-Powered RCA	Root cause analysis workflows

Folders and files

Latest commit

History

Repository files navigation

Vibing Steampunk (vsp)

100 Stars!

What's New

Key Features

Quick Start

Windows 11 with VS Code + Claude Code extension:

1. Get the latest vsp release:

2. Initialize the config files:

3. Adjust your config files:

4. Set up authentication

5. Test the connection:

CLI Coding Agents

CLI Mode

CLI Commands

System Profiles (.vsp.json)

CLI Flags

Environment Variables

.env File

Usage with Claude

Claude Desktop

Claude Code

Transportable Packages Configuration

Focused vs Expert Mode

DSL & Automation

YAML Workflows

Go Library

RAP OData Service Creation

Step-by-Step Workflow

Binding Options

For Transportable Packages

Related

ExecuteABAP

AI-Powered Root Cause Analysis

Tools Reference

Credits

Optional: WebSocket Handler (ZADT_VSP)

Documentation

Development

Project Status

Completed (v2.15.0)

Parked (Needs Further Work)

Planned

Future Considerations

Lua Scripting (New in v2.14)

RCA, Replay & Test Extraction

The Vision: AI-Powered Debugging Pipeline

Example: Capture FM Execution for Test Generation

AI Processing Pipeline

What Works Today (v2.14)

Coming in Future Phases

Related Documentation

Vision & Roadmap

License

Contributing

About

Resources

Uh oh!

Stars

Watchers

Forks

Releases

Packages 0

Uh oh!

Contributors

Uh oh!

Languages

System Profiles (`.vsp.json`)

Packages