RockAI Acknowledgements and Infrastructure Note

An AI-Native Research Infrastructure Partner of the EFT Science Research Laboratory

Core Positioning

The EFT Science Research Laboratory is an open research and experimental organization system built around Energy Filament Theory (EFT). It includes theoretical texts, an AI knowledge base, experiment pages, reproduction materials, multilingual releases, and routes for third-party review. The RockAI team provides AI-native infrastructure support for the lab, helping its materials, records, tasks, and release workflows remain clear, traceable, and maintainable even with a small team.

I. Acknowledgements

The EFT Working Group thanks the RockAI team for supporting the infrastructure development of the EFT Science Research Laboratory. RockAI participates in infrastructure work such as laboratory operations, workflow management, data records, and release coordination.

This support helps EFT Lab organize experimental materials, reproduction indexes, multilingual pages, public communication workflows, and third-party review entry points with greater stability. For a small research team with limited resources, this kind of infrastructure is not itself a scientific conclusion; it is the operating condition that sustains open research, long-term maintenance, and external review.

II. RockAI: AI-Native Workflow Management and Laboratory Operations

Within EFT Lab, RockAI serves as the AI-native workflow management and laboratory operations layer. It works across the organizational relationships among tasks, workflows, people, materials, AI engines, and release nodes, helping the lab break high-density knowledge work into trackable, assignable, and recordable work chains.

In day-to-day work, RockAI supports lab infrastructure setup, workflow organization, material status management, checklist circulation, webpage and knowledge-base release workflows, and internal testing for future forum and collaboration systems. Its role is to reduce workflow fragmentation and make the handoff among research, review, and release clearer.

III. Zulabase: A Data and Run-Record Foundation with AI Capabilities

Zulabase serves as EFT Lab's foundation for data and operational records. Around experiments, reports, and reproduction materials, it stores and connects data sources, run records, parameter ledgers, version states, result indexes, file locations, and publicly released materials.

For EFT Lab, Zulabase functions more like the lab's memory layer: it places the records and materials produced during a research process into a single traceable structure, making it easier to find paths for later review, reproduction, updates, and extended experiments. Its AI capabilities are used mainly for record retrieval, material association, status prompts, and internal management; they do not perform scientific interpretation or make conclusions.

IV. RockAI Agent: A Day-to-Day Workflow Assistant

In EFT Lab, RockAI Agent serves as a day-to-day workflow assistant. Around experiment pages, P-series reports, knowledge-base organization, and multilingual releases, it handles recurring work such as task breakdown, workflow scheduling, material archiving, formatting and link checks, consistency checks between tables and body text, page-generation preparation, and reproduction-material indexing.

RockAI Agent is positioned not as an "AI scientist," but as a laboratory operations assistant. It turns high-context, process-heavy, repetitive work into executable tasks, allowing the research team to keep more attention on scientific thinking, experimental design, data interpretation, and scientific judgment.

V. Infrastructure Contributions in P1

P1_RC_GGL is one of the first relatively complete observational testing workflows publicly presented by EFT Lab. Around P1, the contributions of RockAI, Zulabase, and RockAI Agent are concentrated at the workflow and record layer: task decomposition, experimental-material archiving, data-source and version records, run markers and parameter-ledger organization, pre-release consistency checks for charts, tables, and body text, replication-package indexing, and preparation of experiment and explanation pages for publication.

These efforts help P1-related materials form a clearer chain from data records and run records to reports, reproduction entry points, and webpage publication. They do not determine whether the model is valid, nor do they replace independent reproduction; their value is that they make a complex experimental workflow such as P1 easier to organize, trace, present, and hand over for examination by external teams.

VI. Boundary and Relationship Disclosure

The EFT author is also one of RockAI's co-founders, so RockAI has been prioritized for inclusion in the internal infrastructure development of the EFT Science Research Laboratory. To avoid misunderstanding, we set out the boundaries here:

• RockAI, Zulabase, and RockAI Agent participate in AI-native research infrastructure, data records, workflow organization, and release support;
• The creation of EFT theory, scientific interpretation, experimental design, model judgment, result judgment, and final conclusions remain the responsibility of the EFT author and the EFT Working Group.
• The products and tools described above do not constitute independent evidence for the correctness of EFT, nor do they replace data, experiments, third-party reproduction, public questioning, or peer review.

VII. An Open Stance Toward Third-Party Review

EFT Lab builds this infrastructure so that theoretical texts, experimental reports, reproduction materials, and future research routes can be more easily read, questioned, and reviewed by external teams. We welcome third-party teams to examine the data, review the workflow, reproduce the experiments, propose alternative models, and continue challenging EFT in new observational windows.

RockAI's contribution to EFT Lab is to help this open research environment operate with greater stability; whether EFT holds up will ultimately be judged by data, logic, reproduction, and time.