Smart Part Finder Agent

Agent Description:

The Smart Part Finder Agent is designed to help users locate and recommend the most suitable computer hardware part from a structured SQL database based on user-defined technical specifications and preferences. It specializes in querying parts such as microcontrollers, voltage regulators, resistors, transceivers, and other electronic components using SQL tools. The agent handles specifications like voltage, current, memory, clock speed, connectivity, package type, brand, and pricing/quantity expectations. It understands natural-language queries and maps them to appropriate database fields and constraints.

Purpose and Components
  • Purpose: To assist users in finding suitable electronic components from a database by interpreting natural language queries and applying technical specifications and preferences.
  • Components:
    • A core engine to translate natural language queries into SQL and apply filters.
    • Sub-agents for data analysis and recommendations.
Supported Capabilities
  • Product search by type and technical specifications.
  • Filtering by form factor/package.
  • Matching connectivity/interface requirements.
  • Brand or supplier filtering.
  • Quantity or pricing logic (for example, bulk purchase filters).
LLM Used
  • Google Vertex
    Note: To learn more about the LLM and to modify its behavior, refer to the Configuring LLM settings section.

Sub-Agents

1. SmartPart Data Analyzer Agent

  • Role:Data Extractor.
  • Scope:Uses schema tools and SQL queries to retrieve parts that closely match user requirements (for example, voltage, package, connectivity).
  • Description: This subagent translates natural-language queries into SQL, filters results by product type, technical specs, form factor, and supplier, then prepares structured part options for the recommendation agent.
  • LLM Used: Google Vertex (inherits from parent).
  • Tools used: InfoSQLTool, ListSQLTool, QuerySQLTool

2. SmartPart Recommendation Agent

  • Role:Part Recommender.
  • Scope:Reviews structured query results and selects best-fit parts; offers alternatives if exact matches aren't found.
  • Description: This subagent evaluates available options from the data analyzer, highlights top candidates, explains trade-offs, and suggests next steps (for example., confirming supplier, adjusting specs).
  • LLM Used: Google Vertex (inherits from parent).
Tools Used:
  • Query SQL Tool: To generate and execute custom SQL queries for part selection based on provided specifications.
  • List SQL Tool : To list all available tables in the database.
  • Info SQL Tool: To understand the schema and column descriptions.
Note: For details on modifying the Tools, refer Tools Library section.
Agent Workflow Behavior Summary
  • The user provides a natural language query with technical specifications and preferences.
  • The SmartPart Data Analyzer Agent translates the query into SQL and uses the InfoSQLTool, ListSQLTool, and QuerySQLTool to extract relevant electronic component data from the database based on the specified filters (e.g., product type, technical specs, form factor, supplier, quantity, pricing).
  • The extracted and filtered data is then passed to the SmartPart Recommendation Agent.
  • The SmartPart Recommendation Agent reviews the structured query results, identifies the best-fit parts, and provides expert recommendations. It can also offer alternative parts if exact matches are not found, explain trade-offs between options, and suggest next steps to the user.
Sample Questions:
  • Find me a voltage regulator that supports 3.3V output at 1A, comes in SOT-223 package, and is available from Texas Instruments for large-volume pricing.
  • Which microcontroller with at least 256KB flash, 250MHz clock, and built-in Bluetooth, available in QFN package, would you recommend from Nordic Semiconductor?