Skip to content
Back to skills

Big Research Analysis

ASecurity

Carry out or inspect authorized research analysis, proofs and interpretations with traceable artifacts, uncertainty, counterevidence and method-appropriate limits.

  • 13 stars
  • 0 votes
  • 0 copies
  • 2 views
  • Added October 17, 2026
ai-agents

Works with

  • cli

Security analysis

A100/100

Pro scans all 2 files and shows the line behind each finding

Scanned October 17, 2026

npx -y skills add HaoYangJin/Big-Research --skill big-research-analysis --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Big Research Analysis?

Add the live security badge to your README. It updates with every re-scan.

Security grade badge for Big Research Analysis
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/haoyangjin-big-research-analysis/badge)](https://www.skillsdirectory.com/skills/haoyangjin-big-research-analysis)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
SKILL.md
---
name: big-research-analysis
description: Carry out or inspect authorized research analysis, proofs and interpretations with traceable artifacts, uncertainty, counterevidence and method-appropriate limits.
---

# Investigate and interpret

Read the protocol, selected methods, actual materials and available tools. Record exploratory changes; do not silently alter confirmatory comparisons after seeing outcomes. Do not execute uninspected external code or assume a recorded command was run.

Use authorized client execution tools for actual programs; the workspace only records outputs. Save commands, environment, parameters/seeds where applicable, source/configuration, raw outputs and errors. Keep planned, failed and executed runs separate. Preserve negative results without favorable-seed selection.

For quantitative work, inspect independence, missingness, confounding, leakage, matched budgets and uncertainty. For qualitative or interpretive work, trace analysis to excerpts and context, preserve memos, reflexivity and disconfirming cases. For theory, expose proof steps, assumptions and counterexample searches; finite code checks do not establish universal proof.

Separate observations, interpretation and recommendations. Link bounded claims directly to eligible evidence and relevant contradictions. A documented interpretation is not a universal fact. Save actual artifacts, concerns and limits; changed evidence requires renewed review of affected conclusions.

Files in this skill

  • SKILL.md1.4 KB
  • agents/openai.yaml178 B

Attribution

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments

Loading comments…

Newsletter

Get the weekly list of safe new skills, and the ones we pulled.