Skip to main content

Documentation Index

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

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

Chat Home Screen
Chat is Prisme.ai’s enterprise-grade conversational interface that enables all users to interact with AI in a secure, managed environment. Found under the Use menu in the sidebar, Chat has two main entry points:
  • Chat - Your home for starting conversations, attaching files, and reopening recent Canvas work
  • Agents Space - Discover, browse, pin, and start conversations with agents available in your organization
Chat offers a familiar chat experience similar to consumer AI tools, but with the security, privacy, and integration capabilities required by organizations.

Key Features

Multi-LLM Support

Connect to multiple language models from leading providers or use your own models

Document Processing

Attach files to a conversation so agents can summarize, compare, and answer questions from the content

Canvas Collaboration

Collaborative content creation workspace with AI assistance

Multimodal Capabilities

Process and analyze images and audio when using compatible LLMs

Conversation History

Securely save and retrieve past conversations

Enterprise Security

End-to-end encryption, access controls, and compliance features

Knowledge Integration

Connect to knowledge bases for organization-specific information

Prompt Library

Browse curated prompt templates by category to get started faster

Custom Tools

Run agent tools, Canvas, web search, and code interpreter capabilities when enabled by your workspace

Who Uses Chat?

Chat is designed for all users within an organization, from executives to front-line employees. No technical expertise is required to start getting value from the platform.
Knowledge workers use Chat to:
  • Draft and refine documents, emails, and presentations
  • Research information from internal and external sources
  • Summarize lengthy documents and extract key insights
  • Collaborate on content creation using Canvas

Core Capabilities

Conversational Interface

Chat provides a familiar interface where users can ask questions, choose an agent, pick an available model, and continue previous conversations.
Chat Conversation
  • Real-time streaming responses
  • Rich text formatting
  • Code highlighting and code interpreter support when enabled
  • Multi-turn conversations with context retention
  • Conversation search, recent history, sharing, and cleanup controls
  • Suggested prompts from each agent to help users start with the right context

Document Processing

Upload and analyze documents directly within conversations:
Chat File Upload Picker with Neutral Example Documents
  • The Attach file button is available in the composer
  • Support for common business formats such as PDF, Word, Excel, PowerPoint, and images
  • Text extraction and OCR capabilities
  • Document summarization and analysis
  • Question answering based on document content
  • Table, chart, and structured content analysis when supported by the selected model and tools

Canvas Collaboration

Canvas opens beside the conversation when an agent creates or updates an artifact such as a document, report, code file, dashboard, or generated webpage:
Canvas in Chat

For Non-Technical Users

  • Rich content editor with intuitive interface
  • Intelligent context menu for generating responses, translations, and text refinement
  • Preview and edit modes for generated artifacts
  • Export controls for supported artifact types

For Technical Users

  • Advanced code editor with syntax highlighting
  • Specialized context menu for code-related operations
  • Real-time preview for web content
  • Direct export or download of generated code and files

Multimodal Capabilities

When using compatible LLMs, Chat can process and analyze visual content:
Chat Multimodal Image Analysis Conversation
  • Image analysis and description
  • Visual question answering
  • Chart and diagram interpretation
  • Document layout analysis
  • Audio, image, and file processing when enabled by the selected model and workspace tools

Agent Tools

Agents can expose specialized tools that run from the conversation, such as business system lookups, project simulators, web search, or code execution. When a tool is available, Chat shows it in the conversation context and in the Tools control near the composer.
  • Use the agent’s suggested prompts to start with a supported workflow
  • Check which tools ran below the agent response
  • Open generated reports, dashboards, and files in Canvas
  • Change the model when your workspace exposes multiple model options
When a tool requires you to authorize Prisme.ai with an external service (Notion, GitHub, Outlook…), a Connect button appears inline in the conversation. See MCP Connections & OAuth for the full flow and admin setup.

Security and Privacy

Chat is built with enterprise security and privacy requirements in mind:

End-to-End Encryption

All conversations and data are encrypted in transit and at rest

Access Controls

Fine-grained permissions for users and groups

Data Residency

Control where your data is stored and processed

Audit Logging

Comprehensive logs of all system activities

PII Protection

Automatic detection and handling of sensitive information

Compliance Support

Features to support regulatory requirements

Integration with Other Prisme.ai Products

Chat works seamlessly with other Prisme.ai products:
  • Agent Creator - Create and configure agents that appear in Agents Space
  • Knowledges - Access organization-specific knowledge bases through agents
  • Builder - Connect structured data workflows to agents and chat experiences
  • Governe - Apply security policies and access controls
  • Insights - Track usage patterns and effectiveness

Getting Started

1

Access Chat

Log in to your Prisme.ai instance. In the sidebar, find the Use menu and click Chat.
Chat Home Screen
2

Start from the home screen

The Chat home screen shows the prompt box, quick prompt categories, available agents, and recent Canvas work. You can:
  • Start typing in the input box to chat with the default agent
  • Click an agent card to start a conversation with a specific agent
  • Open Recent canvas to continue work on an existing artifact
  • Go to Agents Space to browse all available agents
3

Upload documents

Click the Attach file button to analyze specific files.
Chat File Upload Picker with Neutral Example Documents
Once uploaded, you can ask questions about the document content.
4

Use Canvas for content creation

During a conversation, Canvas opens automatically when creating documents or code. Canvas provides specialized tools for both text and code creation.
Chat Canvas Preview

Next Steps

Explore more detailed guides for Chat features:

Getting Started

Detailed first steps for new users

Agents Space

Discover, browse, pin, and start conversations with agents

Document Handling

Learn how to work with documents effectively

Conversation Management

Organizing and managing your chat history

Canvas

Master the collaborative content creation workspace

Multimodal

Working with images and visual content

Prompt Management

Use ready-made prompt templates to get started faster