{"id":"wren-engine","name":"Wren Engine","homepage":"https://github.com/Canner/wren-engine","repo_url":"https://github.com/Canner/wren-engine","category":"data-analytics","subcategories":["semantic-layer","ai-data-access","mcp-server"],"tags":["semantic-layer","mcp","sql","data-governance","ai-agents","postgres","snowflake","bigquery","databricks","rust","python","fastapi"],"what_it_does":"Wren Engine is a semantic layer that gives AI agents and MCP clients governed, business-context-aware access to databases, translating natural language into precise queries while enforcing role-based access controls across 15+ data sources.","use_cases":["Enabling AI agents to query enterprise databases with business-logic context (e.g., revenue, churn definitions)","BI dashboard automation where AI understands aggregations and business terms","Enforcing data governance and RBAC when AI agents access sensitive data","CRM updates and compliance workflows driven by AI with accurate semantic understanding"],"not_for":["Simple direct SQL access without governance requirements","Non-enterprise or hobby projects that don't need a semantic layer","Teams without data models or business metric definitions already in place"],"best_when":"Your AI agents need to query structured enterprise data with accuracy around business terms like revenue or churn, and you need access controls enforced at the data layer.","avoid_when":"You need a quick direct database connection without governance overhead, or your data sources aren't among the 15+ supported platforms.","alternatives":["dbt-semantic-layer","cube","looker-semantic-layer","atscale"],"af_score":63.5,"security_score":65.0,"reliability_score":null,"package_type":"mcp_server","discovery_source":["github","github_awesome"],"priority":"low","status":"evaluated","version_evaluated":"latest (beta, biweekly releases)","last_evaluated":"2026-03-01T09:50:06.415551+00:00","performance":{"latency_p50_ms":null,"latency_p99_ms":null,"uptime_sla_percent":null,"rate_limits":null,"data_source":"llm_estimated","measured_on":null}}