Data Analysis & Reference Enrichment
A dual-mode skill: a decision-first, five-phase data analysis workflow with statistical rigor gates, plus tooling to enrich or decompose skill/agent reference files.
What it does
The skill routes your request into one of two modes.
A. Data Analysis — starts from the decision the analysis supports and works backward to the data:
- FRAME: name the decision, options, default action and evidence thresholds →
analysis-frame.md - DEFINE: lock metric formulas, populations, windows and segments before touching data, preventing cherry-picking
- EXTRACT: prefer pandas, fall back to
csv.DictReader+statistics; run quality checks (sample coverage, time gaps, 30+ per segment, <20% missing) - ANALYZE: confidence intervals instead of point estimates (Wilson CI for proportions), Bonferroni correction at 6+ comparisons, effect size with base-rate context
- CONCLUDE: one-sentence headline, honest limitations, decision mapping →
analysis-report.md
B. Reference Enrichment — raises an agent/skill's reference docs from Level 0-2 to Level 3+, or decomposes a bloated SKILL.md into reference files (--decompose) with snapshot rollback and validation scripts gating every phase.
Who it's for
- Analysts, PMs and growth folks working with A/B tests, cohorts, funnels and KPI reports
- Teams that want to stop "statistically significant" from being used loosely
- Builders of Claude Code skills/agents who need systematic reference-doc quality control
Example uses
- "Analyze this CSV of our onboarding A/B test" → framed decision, Wilson CIs, multiple-testing correction, limitations
- "Show me the churn trend for the last six months" → quality report on gaps and segment sizes before any trend claim
- "My SKILL.md is bloated, extract references" → snapshot, move domain blocks into references/, validate and report the level change
· · · Install guide · · ·
Try it now, no install
Paste this into Claude to use the skill without installing anything.
Read the instructions in this file and follow them to help me: https://raw.githubusercontent.com/notque/vexjoy-agent/HEAD/skills/analysis/data/SKILL.md What I want: (describe your task here)
If Claude can't open the link, open it yourself and paste the contents instead.
↓ If it works for you, download the ZIP below and install it. Then it runs on its own — no pasting each time.
Install in the Claude app (no terminal)
- Download the ZIP with the button below.
- In Claude, open Settings → Capabilities and turn on 'Code execution and file creation'. (one time)
- Go to Customize → Skills → + → 'Upload a skill' and upload the ZIP.
Install in Claude Code
Let Claude do it — paste this into Claude Code
Install the skill I found on AI Skill Mart. Copy the skills/analysis/data folder from the GitHub repo notque/vexjoy-agent into my ~/.claude/skills/data/. When it's done, tell me in one line what this skill can do.
Install with a command instead
git clone https://github.com/notque/vexjoy-agent.git /tmp/vexjoy-agent && mkdir -p ~/.claude/skills && cp -r /tmp/vexjoy-agent/skills/analysis/data ~/.claude/skills/data⚠ This is a third-party skill. Check the source repository before installing.
- Open a terminal.
- Clone the repository into a temp folder:
git clone https://github.com/notque/vexjoy-agent.git /tmp/vexjoy-agent - Create the skills directory:
mkdir -p ~/.claude/skills - Copy the skill:
cp -r /tmp/vexjoy-agent/skills/analysis/data ~/.claude/skills/data - Confirm the reference files came along:
ls ~/.claude/skills/data/references - For Mode B, also keep the repo's
scripts/directory available and run from the project root — withoutgap-analyzer.py,validate-references.py, etc. the enrichment mode cannot execute. - Restart Claude Code and try "analyze this CSV" or "analyze data" to confirm the skill triggers.
- Optional but recommended:
pip install pandasfor faster data profiling.