{"id":"wmde-wikidatamcp","name":"WikidataMCP","homepage":"https://wd-mcp.wmcloud.org/","repo_url":"https://github.com/wmde/WikidataMCP","category":"ai-ml","subcategories":[],"tags":["mcp","wikidata","wikimedia","knowledge-graph","sparql","llm-agents","tooling"],"what_it_does":"Provides a Model Context Protocol (MCP) server exposing Wikidata exploration/query tools (item/property search, entity statements/values, hierarchy retrieval, and SPARQL execution). A hosted endpoint and local/Docker run instructions are provided.","use_cases":["Agentic workflows that need to look up Wikidata entities from natural language","Generating or verifying knowledge graph facts via Wikidata statements and qualifiers/references","Building entity filtering logic using instance/subclass hierarchy","Running Wikidata Query Service (SPARQL) from an LLM/agent and retrieving results","Enriching downstream applications with Wikidata labels/descriptions (via returned tool outputs)"],"not_for":["Use as a general-purpose Wikidata client without understanding SPARQL/data-model semantics","High-assurance/provenance-critical pipelines unless tool outputs are validated and error handling semantics are confirmed","Workloads requiring explicit tenant isolation or guaranteed data residency (not described)"],"best_when":"When an LLM/agent needs standardized, tool-based access to Wikidata content (especially search + statement retrieval + SPARQL) and can handle paginated/limited result sizes itself.","avoid_when":"When you need guaranteed rate-limit guarantees, formal error-code semantics, authentication requirements, or strong operational guarantees that are not documented in the provided materials.","alternatives":["Direct Wikidata APIs/SPARQL endpoint usage","Wikidata Query Service + client libraries","Community SPARQL client tooling integrated into your own agent framework","Other MCP servers for knowledge bases/knowledge graphs (if available)"],"af_score":58.2,"security_score":41.2,"reliability_score":28.8,"package_type":"mcp_server","discovery_source":["github"],"priority":"high","status":"evaluated","version_evaluated":null,"last_evaluated":"2026-03-30T15:31:22.862031+00:00","interface":{"has_rest_api":true,"has_graphql":false,"has_grpc":false,"has_mcp_server":true,"mcp_server_url":"https://wd-mcp.wmcloud.org/mcp","has_sdk":false,"sdk_languages":[],"openapi_spec_url":"https://wd-mcp.wmcloud.org/docs","webhooks":false},"auth":{"methods":["Not documented in provided README content"],"oauth":false,"scopes":false,"notes":"The README does not describe any authentication mechanism (API keys/OAuth), so auth expectations and protections are unknown from the provided materials."},"pricing":{"model":null,"free_tier_exists":false,"free_tier_limits":null,"paid_tiers":[],"requires_credit_card":false,"estimated_workload_costs":null,"notes":"No pricing or quotas were stated in the provided README content."},"requirements":{"requires_signup":false,"requires_credit_card":false,"domain_verification":false,"data_residency":[],"compliance":[],"min_contract":null},"agent_readiness":{"af_score":58.2,"security_score":41.2,"reliability_score":28.8,"mcp_server_quality":70.0,"documentation_accuracy":65.0,"error_message_quality":0.0,"error_message_notes":null,"auth_complexity":90.0,"rate_limit_clarity":10.0,"tls_enforcement":90.0,"auth_strength":20.0,"scope_granularity":0.0,"dependency_hygiene":55.0,"secret_handling":50.0,"security_notes":"TLS is implied by HTTPS for the hosted endpoints, but explicit enforcement and certificate/policy details are not provided. Authentication/authorization is not documented, so access controls are unknown. The server executes arbitrary SPARQL provided by the caller, which increases the importance of server-side safeguards (query limits, timeouts, and output filtering), but those controls are not described. Dependency hygiene is inferred only from visible dependency list (no CVE/status data provided).","uptime_documented":10.0,"version_stability":35.0,"breaking_changes_history":40.0,"error_recovery":30.0,"idempotency_support":"true","idempotency_notes":"All described tools are read-only (search/retrieve/execute query), so they should be idempotent; however, exact server behavior and side effects are not specified.","pagination_style":"none","retry_guidance_documented":false,"known_agent_gotchas":["`execute_sparql` returns up to K rows as CSV; agents must set K appropriately and avoid assuming full result sets.","`get_statements` excludes qualifiers/references/deprecated values; agents needing full provenance should use `get_statement_values`.","Search functions may depend on vector search availability and fall back to keyword search; results/recall may vary.","The README does not document rate limits, timeouts, or error-code semantics—agents may need to implement conservative retries/backoff after observing failures."]}}