Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.getpg.ai/llms.txt

Use this file to discover all available pages before exploring further.

​
Prerequisites

Before using the Agent, ensure you have:
  • A PG:AI account with Agent access enabled
  • At least one company added to your workspace with Intelligence data
The Agent uses Intelligence data as its foundation. The more enriched an account is, the better the agent’s responses will be.

​
Accessing the Agent

1

Navigate to a company

Navigate to any company in your workspace.
2

Open the Agent tab

Click the Agent tab in the company view.
3

Start chatting

A conversation interface appears. You’re ready to start - no additional setup is required. The agent has automatic access to all Intelligence data for the account you’re viewing.

​
Starting a Conversation

Type a question or request in the message input. The agent will respond using:
  • Intelligence data - the account’s strategic priorities, contacts, tech stack, financials, hiring data, and enrichment history.
  • Web search - real-time web search for current information not yet in the Intelligence profile.
  • Deep research - extended research triggered by specific requests (e.g. “research their recent acquisitions in detail”).
Responses include source citations linking back to the data or web pages used, so you can verify and explore further.

​
Conversation Management

  • Multiple conversations - you can create separate conversation threads per account for different research topics.
  • Conversation history - all conversations persist and are accessible to your team. Previous research is not lost.
  • Conversation memory - the agent remembers context within a conversation. Follow-up questions build on previous answers.

​
Configuration Options

The Agent works out of the box with no configuration. Optional settings include:
  • Model selection - choose between available LLM models for different quality/speed tradeoffs (if enabled for your organisation).
  • Memory settings - control how much conversation history the agent retains for context.