AI Insights Chat

AI Insights Chat

The AI Insights Chat feature allows you to interact with Foglight AI Insights through a conversational interface. Ask questions about your database performance, get insights, and receive recommendations using natural language queries.

Accessing AI Insights Chat

Navigate to Foglight > AI Insights Chat in the left navigation pane to open the chat interface.

Chat Interface Overview

The AI Insights Chat interface provides:

• Chat Area - The main conversation panel where you interact with Foglight AI Insights
• Messages Tab - View individual messages and responses
• Chats Tab - Switch between different chat sessions
• New Chat Button - Start a new conversation
• Query Input Field - Type your questions or requests about database performance
• Category Filters - Pre-defined query categories to help you ask relevant questions:
  • DBA - Database administration queries
  • Structure - Database schema and structure inquiries
  • Additional categories based on your MCP server configuration

Starting a New Chat

1. Click the New Chat button at the top of the interface
2. A greeting message appears with your username: “(username), how can Foglight AI Insights help you today?”
3. You can choose from suggested queries or type your own question in the input field
4. Use the DBA and Structure dropdowns to specify the context for your query

Asking Questions

To ask a question about your database performance:

1. Type your question in the Ask about anything input field
2. Optionally, select a DBA type or database Structure from the dropdown menus to provide context
3. Click Enter or click the send button to submit your query
4. Foglight AI Insights will analyze your query using the configured MCP servers and provide insights

Example Queries

• “Show me the slowest queries across all instances”
• “Which instances need immediate attention?”
• “What are the database structure changes in the last week?”
• “Analyze performance bottlenecks for my production database”
• “Compare performance metrics between the last two days”
• “What are the top blocking sessions?”

Accessing MCP Tools

At the bottom of the chat interface, you’ll see: Type ‘#’ to access MCP tools

This allows you to:

• Reference specific MCP tools and capabilities from your configured servers
• Type # followed by the tool name to access advanced MCP functions
• Chain multiple tool calls for complex analysis

Using MCP Tool References

When you type # in the query input field, available MCP tools from your configured servers will be displayed. Select a tool to include it in your query for more precise analysis.

Chat History

All conversations are automatically saved in the Chats tab on the right side panel.

Managing Chat Sessions

You can:

• View Previous Chats - Access all your previous chat sessions from the Chats tab
• Resume a Conversation - Click on any previous chat to continue the discussion
• Organize Analysis - Keep your analysis history organized for future reference
• Start Fresh - Create a new chat anytime using the New Chat button

Chat Status

• No chats yet - When you first access AI Insights Chat, you’ll see “No chats yet” with a message to start a new chat
• Active Chat - Your current conversation is displayed in the main chat area
• Previous Chats - Saved conversations appear in the Chats list for easy access

Best Practices

Provide Context

• Use the DBA and Structure dropdown filters to give Foglight AI Insights context about your query
• Specify instance names or database names when asking about specific databases

Ask Specific Questions

• Be specific about what information you need (performance metrics, bottlenecks, changes, etc.)
• Ask one primary question per query for better results

Review Suggestions

• Foglight AI Insights may suggest related queries at the start of new chats
• Use these suggestions as a starting point for your analysis

Chain Queries

• Use follow-up questions to dive deeper into insights
• Reference previous responses to build on prior analysis

Limitations and Notes