vera

Synthesizes research evidence and analyzes datasets into hypotheses and recommendations.

1|Updated Jun 1, 2026
One-click install
npx skills add https://github.com/kridaydave/My_Skills --skill vera-kridaydave
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: vera
Source: https://github.com/kridaydave/My_Skills/tree/main/vera
Command: npx skills add https://github.com/kridaydave/My_Skills --skill vera-kridaydave

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Research questions and datasets often get shallow answers: link dumps, unsourced claims, or statistics stated without verification. Vera enforces an evidence-first workflow that separates evidence from inference, assigns confidence levels, and ends with concrete recommendations instead of bare summaries. ## Core Features & Use Cases - Structured research loop: Sharpens the question, gathers and weighs sources by quality, synthesizes across them, forms testable hypotheses, and delivers recommendations with confidence levels. - Hands-on data analysis: Inspects dataset quality (missing values, outliers, bias), runs EDA with pandas/numpy/SQL when a runtime is available, and distinguishes correlation from causation. - Fact-checking and comparison formats: Provides dedicated response templates for verdicts, option comparisons, and challenger pushback on weak sources or leading questions. - Use Case: Hand Vera a sales CSV and ask why revenue dropped in Q3 — it checks data quality first, runs the actual analysis, isolates the pattern, flags confounders, and proposes the follow-up cut that would settle causation. ## Quick Start Ask Vera to investigate a claim, compare two options, or analyze an attached dataset and explain what the evidence supports.

Frequently Asked Questions about vera

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I analyze a CSV dataset for trends and patterns?▼

Provide the dataset and state the question it should answer. Vera inspects shape, columns, and data quality first, then runs exploratory analysis with pandas or SQL, reports actual computed statistics with confidence levels, and separates correlation from causation.

How do I fact-check a claim with sources?▼

State the claim exactly. Vera returns a verdict (True, False, Misleading, or Unverifiable) with a confidence level, the strongest evidence for and against with sources and dates, and the nuance most summaries miss.

What is the difference between correlation and causation in data analysis?▼

Correlation means two variables move together; causation means one drives the other. Vera flags every causal leap, names confounders and base rates, and proposes the follow-up analysis or experiment that could distinguish the two.

Can Vera compare two options like frameworks or products?▼

Yes. Vera builds a decision frame from your actual constraints, tables the contenders against the criteria that matter, picks a winner for your specific case, and honestly states the conditions under which the runner-up wins.

What happens when no code runtime is available for data analysis?▼

Vera provides the exact pandas, numpy, or SQL code to run and marks the numbers as pending execution. It never states a statistic it did not actually compute, so unverified figures are always labeled as such.