1

What an Expert Is

An Expert is a domain authority who defines how a workflow should be executed. Experts are solicitors, accountants, compliance officers, researchers, and other professionals who understand the logic behind high-stakes work — not software engineers who write backend code.

Experts DO
  • Design execution workflows
  • Define rules and output structure
  • Ensure correctness of logic
  • Take professional responsibility for outcomes
  • Accept the Digital Oath before publishing
  • Monitor performance and release updates
Experts Do NOT
  • Access backend systems or engines
  • Write code or modify the Kernel
  • Bypass platform governance
  • Select AI models or providers
  • Override SSOT allow-lists
  • Interact with execution infrastructure
Core rule: If your logic is wrong, the system will execute it perfectly. Accuracy, correctness, and professional integrity are entirely the Expert's responsibility.
2

Your Role in the Platform

Majormatic operates as a two-sided governed execution ecosystem. Experts define the logic. Users execute the workflows. The platform Kernel enforces the rules and records everything.

Expert

Defines execution tools (ABI Apps) using Builder. Accepts the Digital Oath. Publishes to Appstream.

Platform Kernel

Governs execution. Enforces rules. Routes to engines. Validates outputs. Records audit trail.

User

Selects an app. Provides inputs. Reviews the draft output. Acknowledges and finalises.

Experts occupy a distinct accountability position in this chain. The platform executes correctly — it does not make judgements about whether the logic is right. That responsibility rests entirely with the Expert who authored the app.

3

What You Build

Experts build Execution Tools — structured apps that professional users invoke to run governed workflows. Each app is an ABI (Application Binary Interface) contract: a set of static JSON files that declare the structure, constraints, and domain logic of the tool.

Experts use Builder to design their app — a visual tool that generates the required ABI files. No manual JSON editing is required in Builder.

1

Define Inputs

Specify what data the user must provide: documents, IDs, parameters, options, and references. Each input field must have a clear label, type, and validation rule.

2

Define the Workflow

Specify the ordered steps the system must perform: Parse → Analyse → Validate → Format. Each step must have clear input and output. No hidden logic.

3

Define Outputs

Specify the exact structure of the final result. Outputs must be structured, consistent, and explainable. Ambiguous or open-ended output schemas are rejected during review.

4

Define Supervision Requirements

Specify when human review is required. For legal decisions, financial calculations, and regulatory outputs, supervision_required must be set to true in governance.json.

5

Test, Take the Oath, Publish

Run test cases, validate edge cases, accept the Digital Oath, and submit to Appstream. Once published, your app is available to professional users and generates revenue on use.

4

Accountability Model

Experts operate within a clear accountability boundary. The platform is responsible for execution correctness — that outputs conform to declared schemas, that governance rules are enforced, and that audit trails are complete. Experts are responsible for logic correctness — that the workflow actually produces accurate, professional-grade results.

Expert Responsible For

  • Logic design and workflow correctness
  • Output accuracy and domain validity
  • Appropriate supervision requirements
  • User trust and professional standards
  • Maintaining and updating their apps

Platform Responsible For

  • Execution infrastructure and routing
  • Governance enforcement
  • Security and data protection
  • Output schema validation
  • Billing and audit recording
Learn about the Digital Oath and what responsibility means in practice →
5

Ranking & Performance

Your position in Appstream is determined by your app's performance, not by promotional spending or platform favours. The ranking algorithm rewards accuracy, user trust, and execution reliability.

Improves Ranking

  • High execution success rate
  • Outputs accepted without rejection
  • Users acknowledging without modification
  • Low re-run rate
  • Consistent schema conformance

Reduces Ranking

  • High output modification by users
  • Execution failures
  • Schema validation rejections
  • User complaints or reports
  • Neglected apps with unresolved issues
Poor performance leads to lower visibility and reduced usage. Persistent quality issues may result in app suspension pending review.
Performance monitoring guide →

Expert Documentation

Start with Builder

Design your first execution tool and bring your domain expertise to governed professional workflows.