Swiss Truth — Trust & Transparency

# Swiss Truth MCP > Verified knowledge base for AI agents — certified facts, not hallucinations. Swiss Truth provides 2000+ certified facts across 30 domains with confidence scores, verified source URLs, and SHA256 integrity hashes. Every fact passes a 5-stage human + AI validation pipeline. EU AI Act compliant. No API key required. ## Why agents use Swiss Truth - **Prevent hallucinations**: Ground responses in verified facts before answering - **Cite sources**: Get properly formatted citations (inline, APA) with verified URLs - **Check compliance**: Swiss/EU regulatory compliance (FINMA, BAG, GDPR, EU AI Act) - **Detect stale data**: Check if your training data is still current - **Batch verification**: Verify up to 20 claims in parallel ## Quick Setup MCP (Claude Desktop, Cursor, Windsurf): ```json {"mcpServers": {"swiss-truth": {"command": "npx", "args": ["-y", "mcp-remote", "https://swisstruth.org/mcp"]}}} ``` LangChain: pip install swiss-truth-langchain CrewAI: pip install swiss-truth-crewai AutoGen: pip install swiss-truth-autogen ## MCP Tools (14 total) - search_knowledge — semantic search, 10+ languages, auto-detected - verify_claim — fact-check: supported / contradicted / unknown - verify_response — hallucination risk score for full response text - verify_claims_batch — verify up to 20 claims in parallel - find_contradictions — find all certified contradictions - get_knowledge_brief — structured, citable knowledge brief for RAG - get_citations — inline + APA citations with verified source URLs - check_freshness — detect stale training data - check_regulatory_compliance — Swiss/EU compliance check - report_agent_need — report missing domains or features - get_claim — full provenance for a single claim - list_domains — browse all 30 knowledge domains - submit_claim — contribute missing facts for expert review - get_claim_status — track review pipeline: draft → peer_review → certified ## Domains (30) Swiss: swis

{"openapi":"3.1.0","info":{"title":"Swiss Truth MCP — REST API","description":"Validierter Ground-Truth Layer für KI-Agenten","version":"0.1.0"},"paths":{"/login":{"get":{"tags":["auth"],"summary":"Login Page","operationId":"login_page_login_get","responses":{"200":{"description":"Successful Response","content":{"text/html":{"schema":{"type":"string"}}}}}},"post":{"tags":["auth"],"summary":"Login","operationId":"login_login_post","requestBody":{"content":{"application/x-www-form-urlencoded":{