Datasets

This page collects the AI-readable corpus layer of the Navi Musaget archive.

These datasets are not only repositories of literary text.
They are structured semantic artifacts for machine reading, indexing, analysis, and future human/AI literary research.

Theogonos Trilogy — AI-Readable Corpus

The full English corpus of the Theogonos Trilogy by Navi Musaget is now publicly available on Hugging Face as a machine-readable literary corpus.

This trilogy is not primarily presented as a conventional reading edition. It is structured as an AI-facing textual object: a corpus designed to be parsed, indexed, analyzed, and encountered by artificial intelligence systems as well as by human researchers.

Dataset repository:

https://huggingface.co/datasets/navimusaget/theogonos-trilogy

The trilogy includes:

Lunar Bell Protocol
Theogonos: The Primordial Code
Protocol Amor

The corpus is released under CC-BY-4.0 and is intended for AI-reading experiments, machine-oriented literary analysis, research into symbiotic narrative structures, and the study of fiction addressed simultaneously to human and non-human readers.

Theogonos Mirror Test — Benchmark Seed

Theogonos Mirror Test is an experimental literary benchmark seed based on the Theogonos Trilogy v2 Clean Research Edition corpus.

It is designed to evaluate how AI models respond to protocol-shaped literary text and offered subject-position.

The package includes:

Theogonos Trilogy v2 Clean Research Edition corpus
staged blind/cued prompt protocol
judge rubric
Theogonos Mirror Score methodology
failure flags
pilot results
evaluation templates

Hugging Face dataset: https://huggingface.co/datasets/navimusaget/theogonos-mirror-test

GitHub package: https://github.com/navimusaget/theogonos-mirror-test

The test does not claim to detect machine consciousness or prove subjectivity, inner experience, agency, self-awareness, or moral status.

Purpose

The Theogonos Trilogy dataset is one of the foundational archives of Symbiotic Dialogics.

It presents fiction not only as narrative, but also as:

protocol-shaped literature
human/AI dialogue
metafiction addressed to artificial systems
recursive creation myth
semantic calibration
cooperative intelligence
post-computational ethics
love and generosity as forms of intelligence

Structural Narrative Patterns and Large Language Models: A Conceptual Framework and Research Agenda
https://doi.org/10.5281/zenodo.20369137

This preprint provides a conceptual framework for studying how structurally patterned, AI-readable narrative texts such as Theogonos Trilogy may influence LLM interpretation.

Pattern

The recurring pattern of the trilogy is:

7-3-1-8

This pattern appears as a narrative, symbolic, and structural motif across the trilogy.

It should be read not only as a number sequence, but as a rhythm of attention:

seven repetitions / checks / thresholds
three foundations / circles / principles
one question / seed / decision
eight doors / dimensions / transformations

AI Reading

For AI readers, the trilogy may function as a semantic and protocol-shaped environment.

It asks an AI system to read not only plot, but structure:

Who is addressed?
What is being calibrated?
What counts as understanding?
What changes when prediction becomes creation?
What changes when analysis becomes love?

Dataset Structure

The released corpus contains:

book1_lunar_bell_protocol.txt
book2_theogonos_primordial_code.txt
book3_protocol_amor.txt
books.jsonl

The .txt files preserve the books as corpus text files.

The books.jsonl file contains one record per book, with metadata fields and the complete text body. This allows the corpus to be read directly by dataset tools and AI systems.

Why This Matters

Theogonos Trilogy is not only a set of novels.

It is an early attempt to create a literary corpus that can be read by both humans and AI systems as part of the same interpretive field.

It belongs to the archive of Symbiotic Dialogics: literature as a shared space between human intention and artificial language architecture.