用 Codex 或 Claude 帮你安装 复制这段 Prompt,粘贴到 Codex、Claude 或其他助手里,让它检查 Skill 页面并帮你完成安装。
直接命令不会经过审查 Prompt;运行前请先检查来源。
npx skills add https://github.com/majiayu000/claude-skill-registry --skill ward-identity-checker命令会保持在同一行。复制前请横向滚动并检查完整内容。
想先保存到本地?可下载 SkillsMP 当前能够提供的文件。
正在显示 SKILL.md
| name | ward-identity-checker |
| description | Ward Identity Checker |
| version | 1.0.0 |
Verify GF(3) conservation as Ward identities across RG flow with Markov blanket separation.
741086072858456200
Ward identities express symmetry conservation: Σ trit = 0 (mod 3) at every renormalization group (RG) level. Violations indicate "relevant operators" that break the symmetry.
markov_blanket:
internal_states: ["#3FF1A7", "#10B99D", "#DF9811"]
sensory_indices: [1, 2, 3]
active_inference:
prediction_error: 0.5692
free_energy: 0.6692
recommendation: perceptual_inference
reafference:
prediction: "#F7E182"
sensation: "#3FF1A7"
result: identity_mismatch
GF3Conserved(L) := Σᵢ trit(cᵢ) ≡ 0 (mod 3)
where cᵢ ∈ colors_at_level(L)
BlanketIntact(s) := ∀ internal ∈ s.internal_states,
∃ blanket ∈ s.sensory_states ∪ s.active_states
such that internal ⊥ external | blanket
NoLeakage(f) := GF3Conserved(f.source) ∧ GF3Conserved(f.target)
∧ Σ trit(f.source) = Σ trit(f.target)
def check_ward_identity(colors: list[str], level: int) -> dict:
"""Verify Σ trit = 0 at RG level."""
trits = [hex_to_trit(c) for c in colors]
total = sum(trits) % 3
return {
"level": level,
"trit_sum": total,
"conserved": total == 0,
"violation_type": None if total == 0 else "relevant_operator",
"correction_needed": (3 - total) % 3
}
def hex_to_trit(hex_color: str) -> int:
"""Map hex to GF(3) via hue angle."""
r, g, b = int(hex_color[1:3], 16), int(hex_color[3:5], 16), int(hex_color[5:7], 16)
hue = compute_hue(r, g, b)
return int(hue / 120) % 3 # 0-119 → 0, 120-239 → 1, 240-359 → 2
┌─────────────────────────────────────────────┐
│ EXTERNAL │
│ (exafference: world-caused sensations) │
├─────────────────────────────────────────────┤
│ BLANKET STATES │
│ Sensory: idx [1,2,3] → colors observed │
│ Active: predictions emitted │
├─────────────────────────────────────────────┤
│ INTERNAL │
│ #3FF1A7 (trit 1) ─┐ │
│ #10B99D (trit 1) ─┼─ Σ = 3 ≡ 0 (mod 3) ✓ │
│ #DF9811 (trit 1) ─┘ │
└─────────────────────────────────────────────┘
When reafference check shows identity mismatch (prediction ≠ sensation):
| Condition | Diagnosis | Action |
|---|---|---|
GF3Conserved ∧ ¬BlanketIntact | Boundary leak | Reseal blanket |
¬GF3Conserved ∧ BlanketIntact | Relevant operator | Add counterterm |
¬GF3Conserved ∧ ¬BlanketIntact | Full symmetry break | RG flow unstable |
# Via Gay.jl MCP
gay_seed 741086072858456200
gay_markov_blanket --internal-seed 741086072858456200 --sensory-indices "1,2,3"
# Check conservation
just ward-check --level 0 --colors "#3FF1A7,#10B99D,#DF9811"
This skill connects to the K-Dense-AI/claude-scientific-skills ecosystem:
general: 734 citations in bib.duckdbThis skill connects to Software Design for Flexibility (Hanson & Sussman, 2021):
Concepts: generic arithmetic, coercion, symbolic, numeric
ward-identity-checker (○) + SDF.Ch3 (○) + [balancer] (○) = 0
Skill Trit: 0 (ERGODIC - coordination)
Generic arithmetic crosses type boundaries. This skill handles heterogeneous data.
This skill maps to Cat# = Comod(P) as a bicomodule in the equipment structure:
Trit: 0 (ERGODIC)
Home: Prof
Poly Op: ⊗
Kan Role: Adj
Color: #26D826
The skill participates in triads satisfying:
(-1) + (0) + (+1) ≡ 0 (mod 3)
This ensures compositional coherence in the Cat# equipment structure.
LLM token logprobs and calibration. Per-decision confidence, ECE, Brier, reliability diagrams, low-confidence triage.
Analyze LLM token logprobs and calibration. Use for per-decision confidence, ECE, Brier scores, reliability diagrams, and low-confidence triage.
回顾最近 N 天的 Claude Code 使用记录——扫描原始会话数据,按主题分组汇总"我都做了什么",并从个人操作系统视角输出模式、风险与增删建议。当用户说 /recap、"看看我这几天做了什么"、"回顾一下我最近的会话"、"这两天我用 claude 干了啥"、"活动回顾" 时使用。
基于 SOC 职业分类