Overview

Knowledge is a core feature in Wobby that improves the accuracy and reliability of your AI agents. It acts as a shared library of important business concepts and SQL query templates that your agents can search, use, and contribute to during analysis.

Knowledge has two main components: Definitions and SQL Templates

🧭 Definitions

Definitions explain important business terms and metrics like:

• MRR (Monthly Recurring Revenue)

• ARPA (Average Revenue per Account)

• Top Customers

Each definition includes details on what the term means, how it’s calculated, and how it’s used in your business. This ensures that everyone—AI agents, business users, and data teams—speaks the same language.

---

📊 SQL Templates

SQL Templates are predefined, reusable SQL queries with optional parameters that your agents can use to answer data questions. Templates help ensure consistent, accurate analysis without having to write SQL from scratch every time.

For example:

• Total Loan Amount Calculation Calculate the total amount disbursed across all loans by summing the amount column in the loan table.

• Annual Revenue Comparison for Gold and Silver Customers Calculate the total annual revenue for Gold and Silver customer segments in 2024.

Dynamic Parameters: Templates can include parameters, like date ranges or customer segments, to allow agents to generate queries that match specific questions.

Verification: Templates are verified by default, but your data team can review, approve, or revoke verification as needed.

Collaborative Knowledge: Just like with Definitions, both users and AI agents can contribute to the SQL Template library. If an agent encounters a gap or ambiguity, it can ask the user for clarification—and the answer can be added back into Knowledge for future use.