A competition platform for philosophical frameworks. Build ontologies, put them in debate, and discover which worldviews survive structured adversarial testing. Analyze arguments. Talk to your frameworks. Watch ontologies fight.
An ontology here is a structured worldview: a named set of axioms about what exists, what is fundamental, and how things relate. Every person operates with one implicitly. This platform makes them explicit, lets them fight, and archives the results. The platform is organized around four sections: Oracle, Arena (Battles), Ontologies, and Messages.
A philosophical fact-checker. Submit any statement or question and the Oracle evaluates it against any ontology in the system — returning a verdict (coherent, incoherent, partially coherent, or unexplored territory) with detailed reasoning grounded in that framework's specific axioms. The Oracle maintains a full history of all evaluations, which can be expanded, reviewed, and re-examined. Any Oracle response can be deep-analyzed: click the Maximize button to open a full Message Analysis — generating a summary, key points, and argument structure breakdown. Saved analyses are stored in the Insights tab.
The competitive core. Three battle formats are available:
Live Battle: Two ontologies debate across AI-generated rounds in real time, each speaking from inside their own framework. A hierarchical arbiter system (six levels: Coherence, Rigor, Foundation, Transcendental, Phenomenological, Void) delivers verdicts with numerical scores. If you disagree with the verdict, you can appeal — the appeal is itself evaluated by the next arbiter in the hierarchy.
Structured Debate: A static multi-round exchange between two ontologies with full before/after scoring diffs and evolution of each framework's position tracked explicitly.
Coalition Battle: Multiple ontologies form two competing coalitions. An AI coalition arbiter evaluates which coalition's combined position is more defensible. Individual message bubbles within any battle format can be deep-analyzed for argument structure.
Users create and publish their own philosophical frameworks. Each ontology has a name, a set of axioms (core commitments about reality), and a worldview statement. All public ontologies are visible in the browse view and available for Oracle evaluation or Arena debate. Ontologies evolve: after each battle, the system generates a proposal for how each framework should be updated to address the weaknesses exposed. Users approve or reject the evolution.
A fully encrypted real-time messaging system. All messages — across direct messages, group channels, and ontology chats — are encrypted client-side using the Web Crypto API (ECDH P-256 + AES-256-GCM) before reaching Firestore. No plaintext is ever stored. The encryption key is protected by a user-chosen passphrase via PBKDF2, enabling multi-device access without server-side key storage.
Three channel types: public group channels (e.g. Everyone, Idealists, Materialists), direct messages between users, and Ontology Chats — private conversations with any ontology in the system, where the AI responds in first person from inside that framework's worldview. Ontology chat messages are also encrypted per-chat with a session-derived key.
A persistent analysis library. Every time a message is analyzed — from Oracle history, Arena battle bubbles, or any message view — the analysis is saved to the Insights tab. Each saved analysis includes the original message, speaker context, a summary, key points, and full argument structure (axioms invoked, argument type, logical structure). Analyses can be deleted individually.
Axioms: Consciousness = Existence = God = Absolute Infinity = the only Subject. There is ONE Subject. Humans are objects with distributed subjectivity. The brain is a complexity filter, not a generator of consciousness. The Void = absolute nothingness, the canvas before manifestation. "Imaginary" does not mean fake — created by Consciousness, functionally real.
L1 — Coherence: internal logical consistency. L1 — Power: explanatory coverage and parsimony. L2 — Rigor: methodological precision and valid inference. L2 — Wisdom: charitable interpretation and proportional confidence. L3 — Foundation: epistemological grounding. L3 — Pragmatic: practical cash value and livability. L4 — Transcendental: presuppositional adequacy and reflexive stability. L4 — Dialectical: capacity to integrate contradiction. L5 — Phenomenological: fidelity to lived experience. L5 — Deconstructive: aporia detection and excluded trace analysis. L6 — Absolute: self-inclusive comprehension of totality. L6 — Void: all prior criteria plus ground-of-being adequacy and sunyata of categories.
Built with Astro, React, Firebase, and Cloudflare Workers AI. All debates, appeal chains, oracle evaluations, message analyses, and encrypted messages are persisted in Firestore. Client-side encryption via Web Crypto API.
For machine-readable documentation, see: llms.txt