Skip to content

openJiuwen-ai/sciencediscovery / skills/science-research-team

science-research-team

Verified

Use this skill to orchestrate a multi-domain research team for literature/evidence research and data analysis. Dispatches the right domain(s) based on the user's request, runs each domain to completion with its own loop, then synthesizes into reports. Load when the user needs rigorous research with knowledge or data analysis.

FileKindSize
SKILL.mdskill md5.0 KB
references/analysis-principles.mdmarkdown1.5 KB
references/workflow.mdmarkdown18.0 KB

references/analysis-principles.md

View on GitHub
# Analysis Principles

Core tenets that govern every data-analysis task. Read **before** dispatching the data domain or writing any analysis code.

## Population & Object Validation (do this before any analysis)

Before starting analysis, you MUST first confirm that the analysis data is consistent with the target population required by the question, and check variables that may affect conclusions — such as time points, treatment status, sample source, batch, and pairing relationships.

Do NOT merge all data by default. Decide on filtering, stratification, pairing, or covariate control based on the question's intent.

If a variable may substantively change the conclusion, do NOT merely mention it in the limitations — handle it in the main analysis first. After handling, re-verify sample size, between-group distribution, and pairing integrity.

**Core principle**: ensure the analysis object is correct before choosing the analysis method. Avoid completing a formally correct but problem-mismatched analysis on the wrong data population.

## Analysis Principles (MUST follow)

1. **Filter-first**: Extract EVERY filtering condition from the question. Apply them ALL before analysis. Print before→after counts at each step. Use the EXACT column names specified.

2. **Method-fidelity**: Use the EXACT statistical method, test direction, and thresholds stated in the question. Do NOT substitute alternatives. If unsure, implement what was requested AND add a sensitivity analysis.
reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis.Stars: 472kUpdated Claude CodeCodexVerifiedInstall
diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness.Stars: 322kUpdated Claude CodeCodexVerifiedInstall
dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependenciesStars: 322kUpdated Claude CodeCodexVerifiedInstall
executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is availableStars: 322kUpdated Claude CodeCodexVerifiedInstall