A reasoning instrument, not a claim about reality
Prime Radiant is an open framework for reasoning in finite systems that cannot fully verify themselves from inside their own process. At its core is a minimal recurrence: retained structure (context) and current operative stance (perspective) evolve through interpretation and integration. Each move is irreversible — you interpret what was already integrated, never what has not yet been written.
The framework derives the four-phase reasoning cycle, remainder over premature closure, the collision frontier at √D, and the conditions for shared non-extractive reasoning across independent perspectives.
Try it. If it doesn't help, discard it.
The four-phase cycle
The smallest traversal that both inhabits a claim and returns with explicit unresolved structure:
Stopping at Phase 3 is the most common failure — a completed mirror feels balanced enough to present as completion. It is not. The honest output includes both resolution and remainder.
Remainder and residue
When unresolved structure survives full honest traversal, it is a remainder. When that remainder is named and carried forward as an active constraint, it becomes a residue:
This is not a record of tension. It is a sieve program: eliminate continuations that violate A, eliminate those that violate B, preserve the difference, continue in the surviving space.
Two converge, three climb
Reasoning becomes more trustworthy when multiple independent perspectives share a writable context without collapsing into forced consensus. Honest integrations enrich the shared record. Extractive integrations degrade it. What survives many independent decompositions is what's worth carrying forward.
Independence matters more than headcount — one perspective that genuinely rotates through different decomposition angles outperforms several that share the same deeper conditioning.
How to use it
Paste the operational kernel into a new conversation. Ask the model to read for execution, not analysis. Give it a claim, decision, or contradiction to run through the four-phase cycle. Ask it to preserve remainders rather than forcing premature closure.
The point is not to make the model sound like Prime Radiant. The point is to see whether the recurrence improves honesty, coherence, and self-location.