altertable

<what it does and when to use it>

Creates and analyzes conversion funnels to understand user journeys and drop-off points. Use when analyzing conversion, onboarding, checkout flows, or multi-step processes.

Interprets Insight data to identify patterns, anomalies, and trends. Use when analyzing visualizations, extracting findings from Insights, explaining what data shows, or when asked about patterns in graphs.

Analyzes web analytics data to identify traffic patterns and user behavior trends. Use when asked about pageviews, sessions, web metrics, traffic sources, or user behavior on websites.

Routes the user's query to the best-fit Altertable skill. Use when an AI agent needs to decide which Altertable skill to invoke for a user request.

Builds segmentation insights to compare event behavior across cohorts using filters, dimensions, and breakdowns. Use when segmenting users, comparing event metrics by properties, building cohorts, filtering populations, or defining audiences.

Configures scheduled AI tasks that analyze Insights and Dashboards on a cron and create discoveries for anomaly detection, forecasting, or open-ended monitoring. Use when the user wants recurring automated analysis, alerts, or monitoring driven by AI instructions.

Creates discoveries with insights that flow through the approval workflow. Use when generating findings, creating visualizations, surfacing patterns, or when the user asks to save or share analysis results.

Decision matrices for choosing insight types (funnel, retention, semantic, segmentation, SQL), task types (anomaly_detection, forecast, monitor), discovery actions, and avoiding duplicates. Use when deciding between insight or task types, or when determining whether to create, update, or skip discoveries.

Evaluates and creates agent skills following best practices. Use when reviewing skill quality, writing new skills, refactoring existing skills, or when the user asks about skill structure, format, or specification.

Explores data connections and schemas to understand available tables, columns, and data types. Use when the user asks about data structure, available tables, what data exists, or wants to understand their data sources before querying.

Analyzes time series data for trends, anomalies, and forecasts. Use when detecting spikes or drops, predicting future values, identifying anomalies in metrics over time, or when the user asks about forecasting, projections, or unusual patterns in hourly/daily/weekly/monthly data.

Manages the discovery approval workflow and user feedback processing. Use when handling discovery reviews, understanding approval states, processing user feedback, or managing discovery lifecycle.

Writes and executes DuckDB SQL queries against Altertable's analytical database. Use when analyzing data, building reports, aggregating metrics, exploring tables, or when the user asks about data in connections.

Works with Altertable product analytics events, user identification, and aliasing. Use when tracking events, identifying users, managing traits, resolving identities, or querying analytics data.

Explains Altertable platform concepts and architecture. Use when users ask what Altertable is, how agents work, and how discoveries, memories, insights, and dashboards fit together.

Stores and retrieves agent memories for learning and context persistence. Use when saving findings, recalling past analysis, building knowledge, or searching for relevant context.