AI Analyst Plus

An AI-powered product analyst that lives inside Claude Code. Ask business questions in plain English, get validated analyses, branded charts, and stakeholder-ready slide decks — in minutes, not days.

18 agents | 55 skills | 40 helper modules | DAG-based parallel execution | PDF + HTML + Google Docs export

---

What This Is

AI Analyst Plus turns Claude Code into a product analytics system. It combines:

• Skills (.claude/skills/) — standards Claude follows automatically (chart styling, data validation, question framing, stakeholder communication)
• Agents (agents/) — multi-step analytical workflows orchestrated by a DAG engine (question framing → data exploration → analysis → storytelling → deck creation)
• Helpers (helpers/) — Python modules for charting, SQL validation, data connectivity, statistical tests, and more
• Knowledge (.knowledge/) — persistent memory for dataset context, corrections, proven SQL patterns, and business glossary

You interact by talking to Claude. Ask a question, invoke a slash command, or let Claude figure out which skills and agents to apply.

---

Quick Start

1. Clone and install

git clone https://github.com/ai-analyst-lab/ai-analyst-plus.git
cd ai-analyst-plus
pip install -e ".[dev]"

2. Connect your data (optional)

In Claude Code, run /connect-data to wire up a warehouse (DuckDB, Postgres, BigQuery, Snowflake) or a CSV directory. Curated public datasets with guides live in data/examples/.

3. Start Claude Code

claude

4. Ask a question

What's our conversion rate by device?

Or run the full pipeline:

/run-pipeline question="Why is conversion dropping on mobile?"

---

Don’t Know What to Do? Just Ask.

Claude knows the entire system — every agent, skill, command, and dataset. If you’re stuck:

What can I do with this data?
What should I run to refresh the deck?
How do I connect my own CSV files?
Which agents handle root cause analysis?
Re-run just the chart maker and deck creator.

You don’t need to memorize anything in this README. Think of it as a reference — Claude is the guide.

---

Five Things You Can Do

1. Ask a quick question

What's our conversion rate by device?

Claude queries the data and returns an answer with a chart. Simple questions get answered in under 2 minutes without running the full pipeline.

2. Run a full analysis

/run-pipeline question="What's driving the decline in conversion?"