JK-asthetic

Zscaler Vantagedata — Security skill for Claude Code

Security community

AI-Powered Multi-Table Data Analyst Agent with deterministic Pandas execution, dual-pass grounding audit, interactive Recharts visualizations, and an n8n-style workflow studio.

How to install Zscaler Vantagedata

This entry records only its repository, not the path inside it, so there is no exact command to give. Open JK-asthetic/Zscaler---Vantagedata and copy the folder into ~/.claude/skills/, or the file into ~/.claude/agents/.

What Zscaler Vantagedata does

AI-Powered Multi-Table Data Analyst Agent with deterministic Pandas execution, dual-pass grounding audit, interactive Recharts visualizations, and an n8n-style workflow studio. Built with Next.js 14, FastAPI & Claude.

Alternatives in Security

Imcodes — The IM for agents 960 ★
Node9 Proxy — The Execution Security Layer for the Agentic Era 209 ★
Migration Agent — Plans a token migration (format, tool, naming or tier): chains token-audit and naming-audit, builds the transf 194 ★

README

VantageData v2.0 — AI-Powered Multi-Table Intelligence

**VantageData** is a production-grade enterprise AI Data Analyst agent that enables business users to query, analyze, and visualize complex multi-table datasets using natural language.

Every answer is backed by deterministic Python/Pandas tool execution, verified through an independent dual-pass grounding audit, and visualized with interactive charts.

🌟 Key Capabilities

1. Dynamic Table Registry & File Uploads

In-Memory Table Registry: Pre-loaded with core SaaS tables (customers, subscriptions, usage, support_tickets, revenue_events) with instant runtime indexing.
Dynamic CSV & JSON Uploads: Business users can upload arbitrary datasets at runtime. Uploaded tables inherit identical analytical tooling with zero special-casing.

2. Token-Efficient Analytical Engine (Pandas Powered)

Never Dumps Raw Data: Unlike basic LLM wrappers that flood prompt contexts with 50-row raw JSON dumps, VantageData executes computations directly in Pandas across the entire dataset (whether 7 rows or 100,000 rows).
Specialized Analytics:
- group_by: Category grouping with aggregations (count, sum, avg, min, max).
- value_counts: Categorical frequency distribution and percentage shares.
- summary_statistics: Full column profiling (null counts, unique counts, numeric stats, top values).
- calculate_metric: Domain metrics (total_mrr, total_arr, churn_summary, high_risk_customers).
- query_table: Pinpoint record lookups with column projection (columns) to eliminate token waste.

3. Interactive Visual Chart Engine 📊

Multi-Type Visualizations: Automatically renders interactive Pie, Donut, Bar, and Line charts in chat message bubbles via Recharts.
Live View Switcher: Users can toggle between Pie, Bar, and Line views on the fly.
Interactive Tooltips & KPI Cards: Hover over slices or bars for exact counts