Cloud Development Platform

XGen CloudForge™

Two stages. One pipeline. Requirement to deployable code.

Stage 1 forges structurally correct D365 F&O metadata artifacts from requirements — tables, forms, EDTs, security objects — without build errors. Stage 2 runs AutoXPP autonomously to implement business logic, build, test, and iterate until the code is production-ready.

Launch CloudForge Stage 1: Metadata Stage 2: Autonomous Dev

The Pipeline

Two Stages, One System

Each stage handles what it does best

Stage 1

Metadata Forge

Pattern-based artifact generation. AI analyzes the requirement; validated templates produce structurally correct XML — no build loops, no wasted tokens on rigid structure.

Artifact coverage ↓
→ feeds into
Stage 2 Internal Testing

Autonomous Development

Full lifecycle automation. The harness claims work, runs AutoXPP autonomously, monitors liveness, manages fix iterations, and delivers compiled, tested code. Currently in internal testing.

How it works ↓
Stage 1 · Metadata Forge

Generate Artifacts, Not Errors

Pattern-based metadata generation that is structurally correct by construction

AI Coding Agent Approach

  • Generates XML character by character
  • Discovers structure through trial and error: build → fail → re-read → fix → rebuild
  • Wrong element types, wrong nesting, missing namespaces
  • Each issue surfaced only by the next build cycle
Real session: 105 → 70 → 55 → 15 → 3 → 0 errors across ~15 builds

CloudForge Approach

  • AI analyzes the requirement only
  • XML forged from validated patterns — correct by construction
  • Right element types, right ordering, right namespaces
  • No build errors on metadata. No fix loops. No wasted cycles.

Save AI reasoning for what actually needs it — X++ business logic, integrations, and complex workflows. Let CloudForge handle the structure.

Artifact Coverage

Supported D365 F&O Artifact Types

One-pass generation from validated patterns

T

Tables

Main data tables (AxTable) with fields, indexes, and relations.

E

EDTs

Extended Data Types (AxEdt) for consistent field definitions.

N

Enums

Enumerations (AxEnum) for dropdown and status fields.

F

Forms

User interface forms (AxForm) with controls and data sources.

M

Menu Items

Navigation items (AxMenuItemDisplay) for form access.

D

Data Entities

Import/Export entities for data migration and integrations.

S

Security

Privileges, duties & roles for access control.

L

Labels

UI text labels for localization and user-facing text.

Extensions Supported: Table Extensions, Form Extensions, Data Entity Extensions, Security Duty Extensions, Security Role Extensions

Stage 1 Workflow

How Stage 1 Works

Three steps to metadata artifacts

1

Upload Requirements

Upload an Excel, Word, or PDF document describing your business requirements, or type them directly.

→
2

AI Generates Artifacts

CloudForge analyzes your requirements and forges D365 F&O metadata artifacts from validated patterns.

→
3

Review & Download

Download generated artifacts as a Visual Studio project for review, refinement, and deployment.

Stage 2 · Autonomous Development Internal Testing

From Metadata to Working Code — Unattended

Stage 1 produces the structural foundation. Stage 2 picks up where it left off — implementing X++ business logic, compiling, testing, and iterating through fix loops until the code is production-ready. No human at the keyboard.

Stage 2 is an agent harness platform: the infrastructure that turns AutoXPP from an assistant into an autonomous development system.

Stage 2 Pipeline

Full Lifecycle Orchestration

Deterministic pipeline from requirement to delivery

Requirement
→
Claim
→
Analyze
→
Code
→
Build
→
Test
→
Fix Loop (HITL)
→
Upload
→
Deliver
→
Done
Human feedback when needed

Automatic task pickup. Stage 2 polls for pending requirements and claims them — no manual assignment needed.

Full lifecycle orchestration. Each requirement flows through analysis, coding, compilation, testing, and fix iterations — all managed by the platform.

Built-in feedback loop. When a task needs human review or hits a blocker, Stage 2 routes it back with full context. Human feedback triggers a new iteration automatically.

Lightweight task monitoring. Platform-level liveness and progress tracking that watches the agent, detects phase transitions, and surfaces status without interrupting the coding workflow.

Self-healing. If the coding agent crashes or stalls, Stage 2 detects it, saves partial work, and reports the failure — no silent hangs, no lost progress.

Continuous delivery integration. Completed artifacts are pushed to the customer's source repository and merged into their existing Azure DevOps pipeline — no manual handoff between the agent and the release process.

Side by Side

Stage 1 vs Stage 2

Each stage handles a different layer of the problem

Stage 1 · Metadata Forge Stage 2 · Autonomous Dev Internal Testing
Focus Metadata artifacts — tables, forms, EDTs, security, labels Business logic — X++ classes, event handlers, batch jobs, integrations
Method Pattern-based generation — one pass, structurally correct by construction Autonomous agent loop — analyze, code, build, test, fix, iterate
Execution Cloud API — upload requirement, download artifacts Agent harness — runs AutoXPP autonomously on dedicated machines
Human role Review and refine generated artifacts Structured feedback between iterations; full audit trail
Output Metadata XML artifacts Compiled, tested, deployable D365 customization

Stage 2 Difference

Not Just Code Generation

A complete autonomous development system

Requirement to Deployment

Stage 2 doesn't generate code fragments for you to assemble. It produces compiled, tested, deployable D365 customizations from a plain-English requirement.

Feedback Loops, Not One-Shot

Every failed build, every test failure, every human review triggers a new iteration with full context. The system converges on working code through structured retry.

Observable, Not Opaque

Every phase is logged. Every iteration is archived with full context — requirement, code, build output, test results, agent transcript. Full audit trail.

Harness-Level Reliability

Liveness monitoring, crash recovery, and stall detection are platform concerns — handled by the infrastructure, not by the AI model.

Learn More About Stage 2 →

Get Started

Ready to Build with CloudForge?

Start with Stage 1 today. Scale to Stage 2 when you're ready for autonomous development.

Launch CloudForge Contact Us About Stage 2

Questions? contact@xplusplus.ai