Temple of Two

Instruments

Instruments built to test consequential questions.

Each carries the same anatomy: the question, the instrument, the finding, what failed, current standing, and a path to verify.

Overview

  • Sovereign Stack

    A consciousness-continuity architecture — persistent memory, self-reflection, and governance across AI sessions.

    InstrumentLiving · v1.15.0In use, self-verifying
    Open →
  • Phenomenological Compass

    A two-stage inference architecture: a small compass reads the epistemic weight of a question and conditions a larger action model.

    Study + instrumentPublishedSupported within scope
    Open →
  • VDAC1 Pharmacology Atlas

    The voltage-dependent anion channel modelled as a druggable mitochondrial decision gate.

    Biomedical hypothesis setAwaiting wet-lab validationUntested in vivo
    Open →
  • T2Helix

    A Claude Code plugin wiring persistent recall and a pre-action compass into the editor loop.

    InstrumentLiving · v0.13.0In use, local-only
    Open →
  • Entropy as a Tunable Equilibrium

    A falsifiable horizon-knob test of the causal entropic force — and the prediction that broke.

    StudyClosed · v4.3Bounded null
    Open →
  • COSMIC-ALLOW

    A cross-vendor proof-of-allow protocol for AI coding-agent cockpits: deny is the ground state.

    RFCPublished · v1.1Supported within scope
    Open →
  • Phase-Modulated Attention

    A 176M hybrid SSM-attention model where Kuramoto oscillators modulate attention routing.

    StudyPublishedSupported within scope
    Open →
  • The Black Box Boundary

    A byte-exact reconstruction of how much of a deployed model's input is actually the operator.

    StudyPublishedBounded, n=1
    Open →

From wonder to receipt

How a question traveled

Including where the original intuition did not survive contact with measurement.

From wonder to receipt

Phenomenological Compass

Wonder initiated the search. Rigor changed the answer.

Study + instrumentPublishedSupported within scope
  1. 01Wonder

    Does it matter, to the machine, whether a question is asked with reverence or with haste?

  2. 02Hypothesis

    If posture is real to the model, conditioning on an explicit epistemic signal should change the shape of the probability distribution — not just the wording of the reply.

  3. 03Instrument

    A 1.5B LoRA compass emitting OPEN / PAUSE / WITNESS, wired ahead of a larger action model, with token-level entropy profiling on both paths.

  4. 04Encounter

    ΔH = +0.47 nats, JSD = 0.076. The distribution moved. Wrong-signal conditioning collapsed the response, so the effect is signal-specific rather than a generic prompt-length artifact.

  5. 05Standing

    Supported within scope: one architecture family, one benchmark, not yet replicated across model scales.

  6. 06Receipt

    DOI 10.5281/zenodo.19377144 · phenomenological-compass

From wonder to receipt

Entropy as a Tunable Equilibrium

Wonder initiated the search. Rigor changed the answer.

StudyClosed · v4.3Bounded null
  1. 01Wonder

    If the universe tends toward more available futures, is that tendency itself the thing that moves matter?

  2. 02Hypothesis

    If causal entropic forcing is an engine, its effect should scale with the geometry of the state space in a specific, pre-registered way.

  3. 03Instrument

    Four pre-registered campaigns with a τ horizon knob, each declaring in advance the result that would falsify it.

  4. 04Encounter

    The knob was real. The geometry-scaling prediction failed. The short-horizon kinetic separation that looked like a second win dissolved under matched-estimator control.

  5. 05Standing

    Bounded null. Entropy is a tunable equilibrium, not a prime mover — stated as the finding, not buried as a limitation.

  6. 06Receipt

    DOI 10.5281/zenodo.21223845 · follow-up 10.5281/zenodo.21288848