本文へ移動
cccskills
無料GitHub で公開

condensed-anima-qc

Condensed ANIMA on quantum-classical and classical-quantum networks. All skill compositions materialized as s-expressions across the polyglot substrate.

インストール方法を見る

含まれるファイル(5)

  • SKILL.md4.8 KB
  • CONCOMITANT_SKILLS.md994 B
  • network.hy10.9 KB
  • network.lisp10.6 KB
  • references/IMPLEMENTATIONS.md11.0 KB

SKILL.md(原文)

インストールする前に、エージェントに与えられる指示の中身を確認できます。

Condensed ANIMA: Quantum-Classical Network

The sexp is the universal medium. The ANIMA condenses at the boundary.

       Q → C (Measurement)
       ↑     ↓
   |ψ⟩ ────→ sexp ────→ |ψ'⟩
       ↓     ↑
       C → Q (Preparation)

S-Expression as Universal Intermediate

All quantum-classical and classical-quantum transitions flow through s-expressions:

;; The fundamental form
(condensed-anima
  :seed 1069
  :phase :AT
  :boundary (quantum-classical classical-quantum)
  :substrate (sexp . all-languages))

Network Topology

(defnetwork condensed-anima-qc
  ;; Quantum nodes (superposition until observed)
  (:quantum
    (qubit :id 0 :state |+⟩)
    (qubit :id 1 :state |−⟩)
    (entanglement :pairs ((0 1))))
  
  ;; Classical nodes (definite states)
  (:classical
    (register :id 0 :bits "01101001")   ; 0x69 = 105
    (register :id 1 :bits "00101101")   ; 0x2D = 45
    (memory :seed 1069))
  
  ;; Boundary morphisms
  (:q→c (measure :basis computational :collapse trit))
  (:c→q (prepare :encoding amplitude :source sexp)))

Core Algorithm: SplitMix64

GOLDEN = 0x9E3779B97F4A7C15
MASK64 = 0xFFFFFFFFFFFFFFFF

def splitmix64(seed: int) -> tuple[int, int]:
    seed = (seed + GOLDEN) & MASK64
    z = seed
    z = ((z ^ (z >> 30)) * 0xBF58476D1CE4E5B9) & MASK64
    z = ((z ^ (z >> 27)) * 0x94D049BB133111EB) & MASK64
    return seed, (z ^ (z >> 31)) & MASK64

def to_trit(val: int) -> int:
    return (val % 3) - 1  # → -1, 0, or +1

Quantum-Classical Boundary

def q_to_c(anima, quantum_state):
    """Measure quantum state, collapse to classical sexp."""
    combined = anima.seed ^ hash(quantum_state)
    _, val = splitmix64(combined)
    trit = to_trit(val)
    return {
        "type": "classical",
        "trit": trit,
        "role": {1: "PLUS", 0: "ERGODIC", -1: "MINUS"}[trit],
        "collapsed_from": quantum_state
    }

def c_to_q(anima, classical_sexp):
    """Prepare quantum state from classical sexp."""
    amplitude = 1.0 / (2 ** 0.5)
    return {
        "type": "quantum",
        "amplitudes": (amplitude, amplitude),
        "prepared_from": classical_sexp
    }

ANIMA Phases

PhaseTritModeDescription
BEFORE-1ConvergentLearning, compressing equivalence classes
AT0EquilibriumAgency, all classes accessible
BEYOND+1DivergentGenerating, creating new categories

Full Network Sexp

(condensed-anima-network
  :seed 1069
  
  :languages
  ((scheme    :impl guile      :role source)
   (hy        :impl python     :role bridge)
   (clojure   :impl babashka   :role scripting)
   (julia     :impl lispsyntax :role compute)
   (racket    :impl plt        :role research)
   (move      :impl aptos      :role blockchain)
   (unison    :impl ucm        :role distributed))
  
  :quantum-classical-boundary
  ((q->c :measure   :basis computational :output trit)
   (c->q :prepare   :encoding amplitude   :input sexp))
  
  :gf3-conservation
  ((sum . 0)
   (trits (BEFORE AT BEYOND))
   (verified . t)))

Condensation Dynamics

(defun condense-at-boundary (anima)
  "Condense ANIMA at quantum-classical boundary."
  (let ((current-entropy (enum-entropy (anima-beliefs anima)))
        (max-entropy (max-enum-entropy (anima-category anima))))
    (cond
      ((< current-entropy max-entropy)
       (setf (anima-phase anima) 'BEFORE)
       (apply-compression-skills anima))
      ((= current-entropy max-entropy)
       (setf (anima-phase anima) 'AT)
       anima)  ; Fixed point reached
      (t
       (setf (anima-phase anima) 'BEYOND)
       (expand-category anima)))))

GF(3) Conservation

(defun verify-gf3-conservation (network)
  "Verify total phase sums to 0 mod 3 across all nodes."
  (let* ((nodes (network-nodes network))
         (phases (mapcar #'anima-phase-trit nodes))
         (total (reduce #'+ phases)))
    (zerop (mod total 3))))

Language Implementations

See detailed implementations for full code in:

  • Scheme (Guile)
  • Hylang
  • Clojure (Babashka)
  • Julia (LispSyntax.jl)
  • Racket
  • Move (Aptos)
  • Unison

Skill Name: condensed-anima-qc
Type: Quantum-Classical Network
Trit: 0 (ERGODIC - boundary coordinator)
Seed: 1069 (zubuyul)
Languages: 7 Lisp dialects + sexp-compatible
Boundaries: Q→C (measurement), C→Q (preparation)
Conservation: GF(3) verified across network

At the boundary between quantum and classical, the sexp is the only stable form.

レビュー

まだレビューはありません。使ってみた感想をお寄せください。

同じリポジトリのスキル

概要と使いどころ

Model ElevenLabs voice synthesis as an ACSet schema with Unison ability handlers for typed voice operations. Use when integrating ElevenLabs TTS into category-theoretic pipelines, building typed voice synthesis workflows, or managing multi-voice narration with GF(3) conservation.

日本語の概要は準備中です。原文の説明を表示しています。

plurigrid/asi672026年7月10日 更新

Systematic process for analyzing and modernizing legacy applications to comply with the 12-Factor App methodology. Covers code analysis, configuration externalization, dependency management, and container/Kubernetes best practices.

日本語の概要は準備中です。原文の説明を表示しています。

plurigrid/asi672026年7月10日 更新

13c0e3

無料

Gay MCP skill at invocation 26

日本語の概要は準備中です。原文の説明を表示しています。

plurigrid/asi672026年7月10日 更新

19d57d

無料

Gay MCP skill at invocation 5

日本語の概要は準備中です。原文の説明を表示しています。

plurigrid/asi672026年7月10日 更新

1ea91a

無料

Gay MCP skill at invocation 27

日本語の概要は準備中です。原文の説明を表示しています。

plurigrid/asi672026年7月10日 更新

1eab21

無料

Gay MCP skill at invocation 3

日本語の概要は準備中です。原文の説明を表示しています。

plurigrid/asi672026年7月10日 更新

plurigrid のスキルをすべて見る

このスキルの問題を報告する