yggy

About yggy

yggy is a knowledge graph platform that converts documents and connected systems into structured, queryable graphs. Instead of relying on text chunk retrieval, it extracts entities and relationships, merges duplicates, and links every fact to its source document. The tool is designed for teams and developers who need to query their data through a web app, REST API, or MCP server, with hosting based in the EU.

Review

yggy approaches the "chat with your data" problem differently than most retrieval-augmented generation tools. It builds an actual knowledge graph from uploaded documents and system connections, with no manual schema design required. The graph then becomes accessible through multiple interfaces: a team app, a REST API, and an MCP server for agent integration.

Key Features

• Automatic entity and relation extraction that populates an ontology without manual modeling
• Duplicate merging with a review queue for ambiguous cases, where merged entities become aliases and every fact retains its original source link
• Reversible merges-wrong merges can be undone from the Resolved tab, splitting entities back apart
• Multi-interface access through the web app, REST API, CLI, SDKs, and MCP server
• Multi-tenant architecture with isolated organizations and projects, SSO, scoped API tokens, and an audit log

Pricing and Value

yggy launched with a free tier and a 25% discount for the first three months on paid plans. Specific pricing tiers beyond the launch promotion are not detailed in the available information. The platform includes a free option for getting started, but long-term pricing structure remains undefined in the reference content.

Pros

• Source traceability is built into the core design-every answer points to its origin document, and users can inspect what the system did and why
• Merges are not destructive; merged entities persist as aliases, and the review queue catches uncertain cases before they're committed
• Developer-focused tooling includes a REST API, MCP server, CLI, and SDKs with documentation written for builders
• The graph engine, graphraum, is open source under Apache 2.0, with a WebGL renderer for 2D and 3D graph visualization
• EU hosting and GDPR-first design address data residency requirements for European teams

Cons

• The platform is not currently open source, though a self-hosted Community edition under AGPL-3.0 is listed on the roadmap
• As a newly launched product, it lacks a track record of production use and community feedback beyond the launch cohort
• yggy is not well suited for users who only need simple text chunk retrieval or who don't require structured knowledge graph outputs

yggy fits teams and developers who need structured, queryable knowledge graphs with verifiable source links and don't want to spend time on ontology design or entity matching. It's also relevant for those building AI agents that need to plug into a clean graph via MCP. The tool is less appropriate for quick, lightweight document search where a full knowledge graph would be overkill.