Onyx
Onyx AI provides an open-source enterprise search and AI assistant platform that connects to tools like GitHub, Drive, Slack, and Confluence, delivering accurate, permission-enforced context to both humans and AI agents. The platform supports any LLM, includes an MCP server for external AI tool integration, and processes over one million queries per week across more than 1,000 teams.
- Artificial Intelligence
- Enterprise Software
- AI Agents
- Developer Tools
- Software Only
Funding
Raised to date
$500KRaised to date based on public sources. This may differ from the amount the company actually raised and is based only on what is publicly available on the internet.
Founders
Founder details are not available yet.
Product
Problem
Enterprise teams struggle to get accurate, comprehensive answers from AI because knowledge is scattered across disconnected tools like GitHub, Drive, Slack, and Confluence. Traditional retrieval approaches search sources one at a time, exhausting context windows and missing critical documents, leading to incomplete answers and poor AI response quality.
Solution
Onyx connects to enterprise knowledge sources and provides a unified retrieval layer that surfaces relevant documents with permissions enforced, delivering accurate, fast, and token-efficient context for both humans and AI agents. The platform combines an LLM's internal knowledge with live internet search and internal team knowledge through app connectors, enabling workflows like investigating cost anomalies or answering complex cross-source questions. Onyx supports any LLM of choice, maintains transparency of data and query processing, and provides a developer platform with rich APIs for integrating RAG, agent, and generative AI capabilities into external applications. The system includes an MCP server that gives other AI tools access to team knowledge, and supports building custom agents and actions to automate complex workflows.
Target Audience
Primary customers are enterprises and teams that want to deploy generative AI safely and effectively across their organization, including engineering, operations, and product teams at companies seeking a secure internal AI interface. The platform is also used by developers and organizations wanting to build customized AI applications on top of their enterprise knowledge base.
Features
- Multi-source retrieval across GitHub, Drive, Slack, Confluence, and other enterprise tools with cross-source document correlation and permission enforcement
- MCP server integration enabling external AI tools like Cursor to access team knowledge for better context
- Developer platform with rich APIs for messages, agent creation, LLM management, Actions registration, and user authentication
- Custom agent and Action builder for automating workflows such as investigating cost anomalies across connected systems
- Open-source architecture with deep customizability and an active developer community
- Flexible deployment options including cloud and self-hosted with enterprise-grade security and compliance
- LLM-agnostic design supporting any model provider while maintaining transparency of data processing