Skip to main content

Praman at a Glance

What is Praman?

  • Playwright Plugin — open-source npm package with 199 typed control proxies, OData fixtures, and SAP auth for test automation engineers
  • Claude Code Plugin — 5 AI agents with plan-generate-heal pipeline for autonomous SAP test automation in Claude Code
  • Claude Cowork Plugin — same agent pipeline with chat-first interface for collaborative SAP test automation in Claude Cowork

When to Use?

  • SAP RISE & Cloud migration — validate Fiori apps before and after cutover
  • Upgrade cycles — continuous regression for UI5 patches & S/4HANA updates
  • No documentation — AI agents discover controls from your live system
  • Cost reduction — free open-source, replace Tricentis & Worksoft
  • Agentic automation — plan → generate → heal with zero scripting

Agent-First Design

Business analysts define the process. AI agents — Claude, Copilot, Jules — generate the tests. No scripting required. Agents connect to your live SAP system via Playwright MCP (JSON-RPC) or Playwright CLI (stdin/stdout) — both installed by default.

Praman Plugin

Drop-in Playwright plugin for SAP test automation engineers. 199 typed control proxies, OData fixtures, Fiori Elements support, and SAP auth — just destructure and go.

npm install playwright-praman
Migration from Playwright →

Praman CLI Agents

Token-efficient browser automation for coding agents. Planner, Generator, and Healer agents work via stdin/stdout — optimized for Claude Code, GitHub Copilot, and Cursor.

npx playwright-praman init
CLI Agents Guide →

Praman MCP Agents

Model Context Protocol server for AI agents with structured accessibility snapshots and rich inline feedback. Ideal for VS Code Copilot, Cursor, and JetBrains AI integrations.

npx playwright-praman init
MCP vs CLI Guide →

Claude Code Plugin

5 AI agents with plan-generate-heal pipeline for SAP test automation in Claude Code. 7 mandatory rules, 19 forbidden patterns, 8 skills, and confidence-scored healing — all enforced automatically.

claude plugin add praman-sap-testing
Claude Code Plugin Docs →

Claude Cowork Plugin

Same 5-agent pipeline and compliance rules — available as a Claude Cowork plugin with chat-first interface. Drag-and-drop install for collaborative SAP test automation with your team.

Install .plugin from Cowork chat
Cowork Plugin Docs →

From business process to Playwright test — autonomously

Submit a Signavio flow, a test case description, or a business process in plain language. Praman's AI agents connect to your live SAP system, discover every UI5 control, generate a structured test plan, and deliver production-ready Playwright scripts — covering SAP end-to-end quality from requirement to deployment evidence. No scripting. No selectors. No delays.

① Input

Signavio Business Flow

② Agent Generates

Test Plan

# Create Purchase Order

### Test 1: Navigate to ME21N
1. Open FLP → Purchasing tile
    expect: ME21N app loads

### Test 2: Enter Header Data
1. Set Vendor: 1000 (Müller GmbH)
    expect: Vendor name auto-filled
2. Purch.Org: 1010, Group: 001
    expect: Header accepted

### Test 3: Submit & Verify
1. Click 'Order' → Post PO
    expect: PO# 4500000xxx
2. OData: GET /A_PurchaseOrder
    expect: Status = Open

③ Agent Generates

Playwright Test

// Praman: AI-first SAP testing
import { test } from
  'playwright-praman';

test('verify purchase order',
  async ({ ui5, odata }) => {

  await ui5.control({
    id: 'vendorInput',
    setValue('1000');
  });

  await odata.read(
    '/A_PurchaseOrder'
  );

  await expect(ui5)
    .toHaveValue('Open');
});

Build your Reports that speak to every stakeholder

📋

Reports by Business Role

Each stakeholder sees only what matters to their role — generated from the same test run.

  • CFOFinancial KPIs, compliance score, period-close status
  • ProcurementPO cycle times, vendor SLA, Procure-to-Pay flow
  • SalesOrder-to-Cash duration, AR validation, collection rate
  • BoardExecutive summary, business risk score, test pass rate
📊

Track Transformation

Asset Accounting (FI-AA)

95.4%Accuracy Score
38KAssets Migrated
$2.3BTotal Asset Value
1,234Depreciation Errors
⚠ Review Required
🔥

Business Flow Performance Heatmap

Identifies bottlenecks by business flow — pinpoint which process is slowing your Financial Period Close.

Depreciation Run (F.16)89.4s
CRITICAL
Period Close (F.16/F110)62.1s
REVIEW
Procure-to-Pay45.2s
GOOD
Asset Accounting29.8s
OPTIMAL
🛡️

Deploy with Evidence

Every run produces audit-ready artefacts — timestamped and structured for sign-off.

  • OData response snapshots per transaction
  • Defect video with step-level timestamps
  • Accessibility score (WCAG 2.1)
  • Data integrity diff: ECC vs S/4HANA
  • CISO-approved, OpenTelemetry traced

The Problem. The Solution. The Proof.

Four questions that define why autonomous SAP testing matters.

Why

End to End Quality Matters

Third party apps, Business flows, Test data, UI5 controls, Delay in timeline.

What

Agentic AI Native Quality Automation

Submit a business process or test case — Praman's AI agents autonomously generate test plans, produce Playwright scripts, validate OData integrity, and interact with every UI5 control. From requirement to evidence in minutes, not sprints.

So What

Deploy with evidence

Data Integrity Validation, Business Flow Performance Heatmap in every run, Accessibility Testing, Compliance.

Then What

Tests that write themselves

AI agents write, maintain, and evolve tests through SKILL.md entry points.

What you get

  • 🤖

    Agent-First Architecture

    SKILL.md entry points, capabilities.list() discovery, compliance verification. Planner, Generator & Healer agents via CLI or MCP. Built for Claude, Copilot, Cursor & Jules.

  • 🔧

    199 Typed Control Proxies & 21 Fixtures

    199 UI5 control types across 8 libraries — sap.m, sap.ui.table, sap.ui.comp, sap.uxap, sap.f. OData V2/V4 interception, Fiori Elements, FLP navigation, 6 auth strategies, SmartField handling. Just destructure and go.

  • 🎯

    SAP UI5 Native — LTS 1.108 · 1.120 · 1.136

    Unified ui5= selector engine with pseudo-classes, combinators, and positional selectors. Auto-retry, self-healing, and 3-tier API resolution. Uses UI5's own control APIs, not brittle DOM selectors.

  • 🔍

    Interactive Inspector & Developer Tools

    npx praman inspect — click any element to see UI5 metadata and ranked selectors. npx praman config for resolved configuration. Extension system for custom matchers & fixtures.

  • 📊

    Enterprise Observability

    OpenTelemetry tracing & metrics (OTLP, Jaeger, Azure Monitor exporters), role-based quality reports, pino structured logging, compliance reporters, defect video. Full visibility into every action.

  • Playwright 1.57+ Native

    Extends @playwright/test. Parallel execution, auto-wait, trace viewer, UI mode. Drop-in replacement — existing Playwright tests work unchanged. Mix page.locator() and ui5.control() in the same test.

  • 🛡️

    Enterprise Security & Cross-Platform

    Future-proof with TypeScript 6.x & 5.x support. CISO-approved, zero npm audit vulnerabilities, SBOM + CodeQL scanning. Windows, macOS, Linux. ESM + CJS dual build. See how Praman compares.

Four Pillars of Evidence

“In Ancient India, Pramāṇa was the standard to separate fact from illusion. In the Modern Enterprise, Praman separates a ‘Successful Deployment’ from a ‘Business Disruption.’”
Pratyaksha

Perception

Real-time, deep-object visibility into the S/4HANA Fiori layer.

Anumāna

Inference

Agentic AI that predicts and heals breaks before user impact.

Upamāna

Comparison

Validation between Legacy ECC and New S/4HANA states.

Shabda

Authority

Playwright — the gold standard of modern web automation.

CISO-Approved
Open Source
Customizable
Cross-Platform
TypeScript

Start testing with evidence.

Open source. Your environment. Your rules.

Get StartedView on GitHub