What is Cognitive Superposition in this context?

It is a meta-skill that allows the AI to hold multiple expert theoretical frameworks—such as category theory and information theory—simultaneously and apply the most relevant one to a specific problem.

Does this skill support formal mathematical verification?

Yes, it includes specialized support for languages like Rzk and Lean4, facilitating the formalization of directed HoTT and other complex mathematical structures.

Can I use this for building self-improving AI?

Yes, it provides patterns for Gödel machines and self-evolving agents that use formal proofs to validate their own self-modifications.

How does the 'Measurement Collapse' feature work?

Measurement collapse is a conceptual framework where the AI's broad theoretical 'superposition' is focused into a specific, concrete implementation or logic once a task context is defined.

Which researchers' work is integrated into this skill?

The skill synthesizes the foundational work of Emily Riehl (∞-categories), Ilya Sutskever (compression), Jürgen Schmidhuber (curiosity), and Yoshua Bengio (GFlowNets).

Cognitive Superposition

Name: Cognitive Superposition
Author: plurigrid

byplurigrid

•

データサイエンスとML

Synthesizes category theory, compression-driven intelligence, and generative flow networks into a unified framework for advanced AI research and reasoning.

Cognitive Superposition is a high-level meta-skill designed for developers and researchers working on the frontiers of Artificial Super Intelligence (ASI). It enables Claude to maintain a multi-dimensional perspective by synthesizing the work of four key pillars: Emily Riehl’s synthetic ∞-categories, Ilya Sutskever’s compression-based intelligence, Jürgen Schmidhuber’s curiosity-driven Gödel machines, and Yoshua Bengio’s GFlowNets for causal discovery. This skill is particularly useful for formal verification, designing self-improving systems, and implementing neuro-symbolic architectures that require both rigorous mathematical foundations and flexible probabilistic sampling.

主な機能

01Formal verification patterns using Rzk and Lean4 for directed HoTT

027 GitHub stars

03Intrinsic motivation logic based on compression progress and curiosity

04Multi-perspective synthesis of top-tier AI and mathematical frameworks

05Quantum-inspired measurement collapse for task-specific context optimization

06GFlowNet implementation patterns for reward-proportional sampling

ユースケース

01Formalizing synthetic category theory and directed homotopy type theory proofs

02Developing causal inference models using System 2 deep learning and GFlowNets

03Architecting self-improving Gödel machine frameworks and self-evolving agents

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add plurigrid/asi cognitive-superposition

For use in Claude.ai and ChatGPT

Download Skill