Contract-first scientific computing and human-in-the-loop discovery across MCP, UAG, simulation and interactive runtimes.

GitHub Organization · Public State & Project Site · IP & Evidence Registry · Canonical Architecture

Foldnite / qFoldIT is a scientific mission infrastructure organization turning scientific and industrial objectives into governed, executable and auditable human/AI missions.

qFoldIT is a scientific mission platform that converts research and industrial objectives into governed, executable and auditable human/AI missions.

The platform separates scientific authority from presentation, gameplay, learning and commercial policy:

Scientific / Industrial Objective
        ↓
Mission Contract
        ↓
Capability Profile
        ↓
Mission Router
        ↓
Mission Compiler
        ↓
UAG / Scientific Object Model
        ↓
Runtime Adapter
        ↓
Human + AI Search
        ↓
Submission / Event
        ↓
CAMEO Reconciliation
        ↓
Scientific Validation
        ↓
Evidence / Contribution Record
        ↓
STATE Projection
        ↓
Enterprise Deliverable

The result is not merely a game level. It is a validated scientific workflow that can be rendered as an interactive experience when that is the best interface for the mission.

qFoldIT is designed to operate as a scientific capability fabric for MCP-compatible agents, including Claude-based research workflows.

The correct boundary is:

Claude / Scientific Agent
        ↓
MCP Scientific Tools & Services
        ↓
qFoldIT Mission Contract
        ↓
UAG / Scientific Object Model
        ↓
Mission Compiler / Runtime Adapter
        ↓
Execution
        ↓
Scientific Validator

Claude is an orchestration and reasoning interface. Scientific truth remains external to the conversational layer and is established by the configured scientific validators and customer-authorized reference boundaries.

One scientific representation can be delivered through multiple interaction surfaces:

                    Canonical UAG / Scientific State
                                │
          ┌─────────────────────┼─────────────────────┐
          │                     │                     │
        UEFN                  Unity                 UNIGINE
          │                     │                     │
   UEFN-TOOLBELT         UNITY-TOOLBELT       UNIGINE-TOOLBELT
          │                     │                     │
          └─────────────────────┼─────────────────────┘
                                │
                          WEB-TOOLBELT
                                │
                         STANDALONE APP

Runtime adapters are execution surfaces, not scientific authorities.

qFoldIT's current OpenStructure integration establishes a server-side scientific validation boundary with protected references and native structural scoring.

CORPORATE_APP
      ↓ authenticated server-side request
OPENSTRUCTURE qFoldIT service
      ↓
OpenStructure / `ost`
      ↓
LDDT / local-LDDT / QS-score
      ↓
ValidationResult
      ↓
validation.completed + audit evidence

A failed validator does not become a score by fallback logic. This fail-closed rule is fundamental to the qFoldIT evidence model.

qFoldIT distinguishes:

1. Authoring clock — target creation and mission compilation.
2. Runtime clock — interaction and event capture.
3. Reconciliation clock — external scientific validation and evidence publication.

For UEFN this distinction is essential: live Verse execution is not treated as a direct arbitrary-network scientific client. The intended model is runtime interaction → session publication → external reconciliation → scientific validation.

qFoldIT treats players and agents as search participants rather than as sources of scientific truth.

Mission State
+ Player / Agent State
+ Historical Outcomes
+ Compute State
        ↓
Learning Policy
        ↓
Decision / Coaching
        ↓
Runtime Outcome
        ↓
Validation
        ↓
Evidence
        ↓
Policy Update

The target metric is validated scientific contribution density, not raw playtime.

qFoldIT's mission architecture is designed for:

• protein and peptide design;
• structural biology and molecular modeling;
• drug discovery and ADMET;
• synthetic biology and L-systems;
• materials and crystal search;
• batteries and energy storage;
• semiconductors and AI hardware;
• quantum computing and hybrid optimization;
• aerospace and space science;
• energy, hydrogen and CCUS;
• mining and resource processing;
• industrial engineering and process optimization;
• scientific software and cloud AI.

qFoldIT distinguishes:

• E1 — verified implementation evidence;
• E2 — strong architecture/product evidence;
• E3 — design, roadmap or option evidence;
• EX — third-party/upstream evidence.

Repository size, external ecosystem reach, prospect lists, TAM, gameplay hypotheses and third-party scientific achievements are not automatically qFoldIT-owned value or revenue evidence.

• qFoldIT GitHub organization
• Public project site and STATE
• Canonical Integrated Platform Architecture
• IP, provenance and valuation registry
• UEFN / Fortnite experience
• FoldIT
• MCP standard

Built on the open Model Context Protocol (MCP) standard.

Scientific authority remains with validated scientific services and authorized reference boundaries.

Global Headquarters: Elkridge, Maryland, USA
CEO: qFoldIT CEO
Email: qFoldIT@gmail.com