science-superpowers
v0.1.0by K-Dense-AI · author: K-Dense, Inc.
Computational-science methodology for research agents: question framing, prior-work survey, analysis design, pre-registration, reproducible execution, anomaly investigation, results verification, and red-team review.
What science-superpowers does, in the publisher's words
Science Superpowers is a complete computational-science methodology for your research agents, built on a set of composable skills plus initial instructions that make sure your agent actually uses them. It has zero third-party dependencies: it runs with only your agent harness and a POSIX shell.
> ⭐ If Science Superpowers helps your research, please star this repository. A star helps other scientists and engineers find the project and tells us the methodology is worth expanding. > > Learn more: Introducing Science Superpowers, why we built it, the Iron Law, and the full workflow. Related essays are collected under From the blog. > > Stay up to date: Follow K-Dense on X, LinkedIn, and YouTube for new skills, release announcements, and research workflow demos.
From the project README, punctuation lightly normalized · Open the README on GitHub
- Skills
- 16
- MCP servers
- 0
- Stars
- 341
- License
- MIT
- Repo created
- 2026-05-28
- Last pushed
- 2026-09-13
- Publisher type
- Organization
- Version
- 0.1.0
Links
Skills · 16
- designing-the-analysis
- Use when you have an approved research question and need a concrete analysis plan, before touching outcome data or fitting any model
- dispatching-parallel-investigations
- Use when facing 2+ independent investigations that can proceed without shared state, parallel literature survey, multi-dataset replication, or pre-specified robustness checks
- establishing-feasibility-first
- 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 e…
- executing-analysis
- Use when you have a pre-registered analysis plan to execute inline in this session with review checkpoints, on a platform without subagents
- framing-research-questions
- 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
- investigating-anomalous-results
- 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
- preregistering-analysis
- 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
- receiving-critical-review
- 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
- reporting-and-archiving-findings
- Use when an analysis is complete and verified, and you need to decide how to report it and archive the work for reproducibility
- requesting-red-team-review
- Use after completing an analysis or before reporting a finding, to have a skeptical reviewer attack the conclusion before you believe it
- setting-up-reproducible-analysis
- 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
- subagent-driven-analysis
- Use when executing a pre-registered analysis plan with mostly independent steps in the current session
- surveying-prior-work
- using-science-superpowers
- verifying-results-before-claiming
- writing-science-skills
Descriptions come from the frontmatter of each SKILL.md, punctuation lightly normalized.
Category
Research & Search. Assigned by this directory. Agent Plugins 1.0.0 has no category field, so no manifest declares one.
Keywords
skills · science · research · pre-registration · reproducibility · statistics · data-analysis · workflows
Related by publisher and keywords
Ready-to-use scientific and research Agent Skills for biology, chemistry, medicine, and related workflows.
Deep research and scientific writing skills: publication-ready papers, literature reviews, grant proposals, posters, slides, and clinical reports backed by verified citations.
Agent Skills for small-molecule and protein therapeutics: target validation and human genetics, bioactivity and chemical space, generative design and retrosynthesis, docking, free energy and dynamics, ADMET and PK translation, protein, antibody, degrader and oligonucleotide design, and the clinical and regulatory record.
Nine local-first scientific research Skills for ideation, hypotheses, study design, statistics, uncertainty, writing, peer review, and scholarly evaluation.
Portable multi-agent harness for.agents-based skills and workflows
Evidence-first research workflow plugins for MiniMax Code, literature search, traceable evidence, citation verification, and human-gated research stages.
[](https://agentpluginsdirectory.com/plugins/science-superpowers)✓ Verified . We fetched the manifest from GitHub and checked it against the official Agent Plugins 1.0.0 schema at agent-plugins.org.