icon file Flask Track Docs

AI Integrations

FlaskTrack includes optional AI-powered capabilities designed to accelerate protocol creation, improve organizational knowledge discovery, simplify reporting, and reduce administrative overhead across laboratory and research operations.

AI features are entirely optional and require an organization administrator to configure an AI provider before they become available.

FlaskTrack does not provide AI models or API access by default.

Overview

FlaskTrack AI capabilities are designed to assist users while maintaining human oversight, traceability, and compliance.

Current AI-powered features include:

All AI-generated content should be reviewed before use in operational, scientific, manufacturing, or regulated environments.

AI Provider Configuration

Before any AI features can be used, an administrator must configure an AI provider.

FlaskTrack supports organization-managed AI providers and does not include AI usage as part of the platform itself.

Organizations are responsible for:

If no AI provider is configured, AI functionality remains unavailable throughout the platform.

OpenAI Integration

Organizations may connect FlaskTrack to OpenAI using their own API credentials.

When using OpenAI, organizations are responsible for:

OpenAI processing occurs according to the organization's configured settings and usage.

Ollama Integration

Organizations may connect FlaskTrack to a self-hosted Ollama deployment.

Ollama allows organizations to operate AI models entirely within their own infrastructure.

Organizations are responsible for:

This deployment option is often preferred by organizations with strict privacy, security, or compliance requirements.

AI Security & Privacy

FlaskTrack respects existing permission controls when using AI features.

AI assistants only access information the authenticated user is authorized to view.

Organizations control:

Knowledge Base indexing does not bypass user permissions.

Users cannot retrieve information they would not otherwise be authorized to access.

Workflow Import Assistant

The Workflow Import Assistant converts existing documentation into structured FlaskTrack workflows.

Instead of manually creating protocols step-by-step, users can provide existing documentation and allow the AI assistant to generate an initial workflow draft.

Supported sources may include:

The assistant analyzes the provided documentation and converts it into structured FlaskTrack records.

Protocol Generation

The Workflow Import Assistant can automatically generate:

Generated protocols can be reviewed, edited, and approved before operational use.

This significantly reduces the effort required when migrating existing procedures into FlaskTrack.

Automatic Material Creation

When importing documentation, the AI assistant can identify materials referenced throughout the procedure.

Examples include:

If materials do not already exist within FlaskTrack, draft material records can be generated automatically.

Users may then review, edit, and approve these records.

Automatic Tool Creation

The AI assistant can identify equipment and tools referenced within imported documentation.

Examples include:

Draft tool records can be generated automatically and associated with the appropriate protocol steps.

Catalog Expansion

Workflow imports can automatically assist in building organizational catalogs.

The assistant may generate draft records for:

All generated records remain editable and should be reviewed prior to production use.

AI Documentation Assistant

The AI Documentation Assistant provides conversational access to organizational knowledge stored within FlaskTrack.

Users can ask questions using natural language and receive answers generated from approved organizational content.

Examples include:

How do we perform contamination investigations?

What is our rooting protocol for banana plantlets?

Which workflow is used for plasmid confirmation?

What equipment is required for media preparation?

The assistant helps users quickly locate information without manually searching through multiple records and documents.

Retrieval-Augmented Generation (RAG)

The Documentation Assistant uses Retrieval-Augmented Generation (RAG).

Before generating a response, FlaskTrack retrieves relevant organizational information and supplies that information to the AI model.

This allows responses to be grounded in actual organizational knowledge rather than relying solely on model training data.

Benefits include:

Knowledge Sources

The AI Documentation Assistant can retrieve information from approved organizational content.

Depending on configuration, indexed sources may include:

The assistant uses these sources to provide context-aware responses.

Notes & Knowledge Base Integration

FlaskTrack Notes can be optionally indexed into the Knowledge Base.

Indexed notes become available to AI-powered retrieval systems.

Examples of useful indexed content include:

Organizations control which notes participate in AI retrieval.

Private notes remain private and are not exposed to unauthorized users.

File & Document Intelligence

The Documentation Assistant can analyze uploaded documents and stored files.

Examples include:

This enables users to quickly understand and operationalize large volumes of documentation.

AI SQL Assistant

FlaskTrack includes an AI-powered SQL Assistant designed to simplify report creation.

Users describe the report they want in natural language, and the assistant generates SQL queries against the reporting data model.

Examples:

Show all deviations by month.

Compare contamination rates between workflows.

List protocol completion times by technician.

Show sample throughput by species over the last six months.

The assistant translates business questions into executable reporting queries.

Report Building Assistance

The SQL Assistant helps users create reports without requiring advanced SQL expertise.

The assistant understands the FlaskTrack reporting schema and can assist with:

Generated SQL can be edited, reviewed, and executed directly within the reporting environment.

Data Lake Integration

The AI SQL Assistant operates against FlaskTrack's reporting layer and analytical data lake.

This reporting environment contains operational events that have been transformed into reporting datasets optimized for analysis.

Examples of available reporting data include:

This architecture allows complex reporting without impacting operational workloads.

Human Review Requirements

AI-generated content should always be reviewed before use.

Organizations should verify:

AI is intended to accelerate work, not replace scientific, operational, or compliance review processes.

Benefits

AI capabilities help organizations:

Best Practices

Related Features