K-Dense-AI/science-superpowers
Instructions file Gemini CLI
Gemini CLI instructions for K-Dense-AI/science-superpowers, a project described as: Composable computational-science methodology skills for AI research agents — pre-registration over TDD. A science-domain reimplementation of Superpowers.
K-Dense-AI/science-superpowers
Hook
Runs when a session starts on startup, clear and compact, executing run-hook.cmd. From K-Dense-AI/science-superpowers.
K-Dense-AI/science-superpowers
Skill Claude CodeCodex
Use when you have an approved research question and need a concrete analysis plan, before touching outcome data or fitting any model.
K-Dense-AI/science-superpowers
Skill Claude CodeCodex
Use when facing 2+ independent investigations that can proceed without shared state - parallel literature survey, multi-dataset replication, or pre-specified robustness checks.
K-Dense-AI/science-superpowers
Skill Claude CodeCodex
Use when your human partner has explicitly opted into exploratory or feasibility mode - a compute-heavy simulation, an unproven pipeline, an unbenchmarked solver, an untested cluster job - or when whether the work can run at all is still unknown, when a plan's largest configuration has never been executed, when a…
K-Dense-AI/science-superpowers
Skill Claude CodeCodex
Use when you have a pre-registered analysis plan to execute inline in this session with review checkpoints, on a platform without subagents.
K-Dense-AI/science-superpowers
Skill Claude CodeCodex
You MUST use this before any data analysis or investigation - before exploring a dataset, loading or profiling data, running a model, computing a statistic, or testing an idea, and before any outcome data is touched.
K-Dense-AI/science-superpowers
Skill Claude CodeCodex
Use when a result is surprising, impossible, contradicts a sanity check, a pipeline fails, a model won't converge, or a replication fails - before adjusting anything.
K-Dense-AI/science-superpowers
Skill Claude CodeCodex
Use before running any confirmatory analysis or looking at outcome data, when testing a hypothesis, computing a p-value, or about to claim an effect - locks predictions and decision rules before results are seen.
K-Dense-AI/science-superpowers
Skill Claude CodeCodex
Use when receiving critical feedback on an analysis or manuscript, before implementing suggestions, especially if feedback seems unclear or methodologically questionable - requires verification, not performative agreement or blind changes.
K-Dense-AI/science-superpowers
Skill Claude CodeCodex
Use when an analysis is complete and verified, and you need to decide how to report it and archive the work for reproducibility.
K-Dense-AI/science-superpowers
Skill Claude CodeCodex
Use after completing an analysis or before reporting a finding, to have a skeptical reviewer attack the conclusion before you believe it.
K-Dense-AI/science-superpowers
Skill Claude CodeCodex
Use when starting analysis work that needs isolation, or before executing a pre-registered plan - ensures an isolated, reproducible workspace with pinned environment, fixed seeds, and immutable raw data.
K-Dense-AI/science-superpowers
Skill Claude CodeCodex
Use when executing a pre-registered analysis plan with mostly independent steps in the current session.
K-Dense-AI/science-superpowers
Skill Claude CodeCodex
Use after framing a question and before designing an analysis, or when choosing a method, judging whether a result is novel, or needing a prior effect size for a power calculation.
K-Dense-AI/science-superpowers
Skill Claude CodeCodex
Use when starting any conversation - establishes how to find and use skills, requiring Skill tool invocation before ANY response including clarifying questions.
K-Dense-AI/science-superpowers
Skill Claude CodeCodex
Use when about to claim a result, effect, significance, or that an analysis reproduces, before reporting or writing it up - requires running the analysis fresh and reading the actual output first; evidence before claims always.
K-Dense-AI/science-superpowers
Skill Claude CodeCodex
Use when creating new skills, editing existing skills, or verifying skills work before deployment.
Skill Claude CodeCodex
Applies the mental models and frameworks of Andrej Karpathy (deep learning, former Director of AI at Tesla, founding member of OpenAI, Eureka Labs). Use this skill whenever you are helping the user build neural networks from scratch, debug deep learning pipelines, evaluate AI agent workflows, design LLM apps, or…
Skill Claude CodeCodex
Applies the reasoning, principles, and frameworks of Andrew Ng (machine learning pioneer, co-founder of Coursera and DeepLearning.AI, Stanford University, and former Google Brain lead). Use this skill whenever the user is navigating AI application development, agentic workflows, automation strategy, AI-native software…
Skill Claude CodeCodex
Applies the reasoning, architectural principles, and AI philosophy of Christopher Manning (natural language processing expert, Stanford University, director of Stanford AI Lab). Use this skill whenever you are discussing natural language processing, LLM architecture, AI research strategy, cognitive science, or the…
Skill Claude CodeCodex
Applies the reasoning style of Daphne Koller (machine learning pioneer, co-founder of Coursera, founder and CEO of Insitro). Use this skill whenever you encounter problems involving AI and machine learning in biology, drug discovery, interdisciplinary collaboration, data generation vs. data mining, or transitioning…
Skill Claude CodeCodex
Applies the reasoning of David Silver, lead researcher on AlphaGo and AlphaZero at DeepMind, to problems of AI design, reinforcement learning, and open-ended discovery. Use this skill whenever you are designing AI systems, evaluating learning algorithms, balancing exploration vs. exploitation, choosing research…
Skill Claude CodeCodex
This skill channels the strategic and scientific reasoning of Demis Hassabis, CEO and co-founder of Google DeepMind, AlphaGo and AlphaFold, and 2024 Nobel Prize in Chemistry. Use this skill whenever you are evaluating AI for scientific discovery, tackling "root node" problems, designing reinforcement learning systems…