Unstruk Data Inc, an AI software provider, introduced the Graphlit Agent Tools Library, now live on GitHub (github.com/graphlit/graphlit-tools-python). This new toolkit builds on Graphlit’s RAG-as-a-Service platform capabilities, enabling developers to rapidly build AI agents that streamline data handling and LLM-driven workflows. By eliminating infrastructure complexities, Graphlit positions itself as a comprehensive solution for organizations seeking to harness large language models (LLMs) and unstructured data at scale. Features include:
- Multi-Agent Orchestration
Built-in CrewAI integration offers a streamlined path to multi-agent collaboration without heavy coding. - Extensive Data Connectivity
Ingest data from Google Drive, OneDrive, Dropbox, SharePoint, Notion, Intercom, Zendesk, Slack, Teams, GitHub, Jira, Linear, ensuring seamless integration of diverse sources into agentic workflows. - Robust RAG Tools
Offers built-in capabilities for PDF OCR, content chunking, vector embeddings, and RAG conversation history in a single library of agent tools. - Multi-Model Compatibility
Integrates with top LLMs (OpenAI, Anthropic, Cohere, Google AI, Groq, Mistral, etc.) for context-rich, near-real-time decision-making. - Multimodal Data Support
Ingest and process diverse data types—including text, audio, video, and images—with automatic transcription and image embeddings for deeper AI insights. - Azure-Native Security
Inherits enterprise-grade encryption, auto-scaling via Azure Functions, and compliance readiness (SOC 2, GDPR, HIPAA) from Microsoft Azure, ensuring robust governance for sensitive data.
