Semantica
An MIT-licensed Python toolkit for AI context graphs, with source-linked decision records, rule-based reasoning and agent framework integrations.
Semantica is a Python toolkit for teams building AI systems that need shared context, traceable decisions and policy checks. It connects facts and decisions to their evidence in queryable knowledge graphs.
- Context graphs: Extract entities, relationships and events from multiple sources, flag conflicting facts and merge duplicates while preserving provenance.
- Decision records: Search earlier decisions, trace causal links and inspect downstream effects or point-in-time graph snapshots.
- Rules and governance: Apply deterministic reasoning, validate graph constraints and edit ontologies in a browser workbench.
- Evidence trails: Attach W3C PROV-O provenance to facts and export audit data in formats including JSON, CSV and RDF.
- Agent and data connections: Share context through Agno, CrewAI or LangChain, expose it through MCP or a REST API, and connect enterprise data from Databricks, Snowflake and SAP OData.
Graph construction, reasoning and provenance don't require an LLM. Storage choices include embedded Oxigraph, Neo4j and FalkorDB, alongside vector stores such as FAISS, Qdrant and PgVector. Retrieval combines graph traversal with semantic search.
Semantica is open source under the MIT license. The browser workbench supports graph exploration, ontology editing and temporal views.
