# Vendor facets — WP Engine. Managed WordPress hosting; hosting earns nothing. Its agent # offering is the AI Toolkit's Smart Search MCP, a two-tool (search, fetch) server hosted on # WP Engine's own api.ai.wpengine.com — one shape per customer, on the vendor's domain, so it # grades `platform`. Files the provider writes into the site root earn what the provider's work earns. vendor: wpengine name: WP Engine website: https://wpengine.com areas: - hosting registry_keys: - wpengine rubric_schema_version: 0.22.0 generated: '2026-09-25' features_refreshed: '2026-09-25' basis: capability summary: >- WP Engine moves no Kin Score check by hosting a site. Its AI Toolkit add-on gives a site a Smart Search MCP server, but the server lives on WP Engine's domain with the same two tools for every customer, so it grades `platform` (0.25 of mcp_server). A provider can upload its own llms.txt and .well-known documents into the site's public root over SFTP, and those earn what the provider wrote. features: - id: smart-search-mcp name: Smart Search AI MCP server (AI Toolkit) description: >- A beta MCP server at https://api.ai.wpengine.com/v1/smart-search/{site-name}/mcp exposing search and fetch over the site's vectorised WordPress content, with a bearer token from the User Portal. source: https://developers.wpengine.com/docs/ai-toolkit/smart-search/mcp/ tier: paid - id: ai-toolkit name: AI Toolkit (Smart Search, Managed Vector Database) description: >- Semantic and hybrid site search, a managed vector database and MCP connectivity, sold as a license on Premium and Headless Platform plans; bring your own LLM. source: https://wpengine.com/support/ai-toolkit/ tier: paid - id: sftp-root name: SFTP access to the site's public root description: >- SFTP uploads anywhere within the site's public root file system (nothing above it), so a provider can place static files such as llms.txt or .well-known documents. source: https://wpengine.com/support/sftp/ tier: all maps: - feature: smart-search-mcp check: mcp_server layer: agent_readiness grade: platform note: >- Hosted on WP Engine's domain with an identical search/fetch tool pair for every customer — the vendor's server shape, not the provider's. `platform` (0.25) in 0.22.0. provider_must: Buy the AI Toolkit license and declare the endpoint in its mcp/ manifest. points: 12 baseline_pass_rate: 0.22 - feature: sftp-root check: llms_txt_published layer: composite provider_must: Write the llms.txt and upload it to the site root; WP Engine generates none. catalog_pass_rate: 0.351 facet: discoverability points: 4 baseline_pass_rate: 0.65 - feature: sftp-root check: well_known_published layer: composite provider_must: >- Author an api-catalog linkset, security.txt or protected-resource document and upload it under .well-known/ in the public root. conditional: true condition: >- Only where the site's server configuration serves the uploaded dotted directory with a usable Content-Type — WP Engine's SFTP docs confirm root placement, not how .well-known responses are served. catalog_pass_rate: 0.005 facet: discoverability points: 6 baseline_pass_rate: 0.018 earns_nothing: - feature: ai-toolkit why: Site search and a vector database improve the site; no check reads them. out_of_reach: checks: - contract_present - spec_presence - protected_resource_metadata - agentic_commerce - agent_card note: >- A host serves what the provider builds; WP Engine documents no contract, auth-metadata or commerce-profile tooling for the customer's surface. surface: discoverability: reachable: 10.0 total: 54 agent_readiness: reachable: 3.0 total: 139 hard_rule: >- A model, not a score. Adopting this vendor changes a provider's Kin Score only when the provider publishes the resulting artifacts on its own surface; nothing here writes a score, and no sponsorship or partnership can. method: searched source: - https://developers.wpengine.com/docs/ai-toolkit/smart-search/mcp/ - https://wpengine.com/support/ai-toolkit/ - https://wpengine.com/support/sftp/ measured: cohort: method: vendors-catalog.json detections (CNAME / header / URL shape / markup), never a name match detected: 3 in_baseline: 0 control: basis: providers earning contract_present + documentation_present + api_reference_present, minus the cohort n: 5216 metric: >- cohort_pct / control_pct = mean share of the check's points earned (derived and platform credit weighted), x100 measured_on: '2026-09-25' status: 'not measurable: 0 detected customers clear the baseline (need 20)' simulation: simulated_on: '2026-09-25' rubric: 0.23.0 population: providers publishing a contract (contract_present earned), replayable exactly providers: 8977 providers_unreplayable: 987 providers_moved: 7325 conditional_rows: excluded (they depend on what the API already does) composite_lift: median: 0.0 p75: 0.7 p90: 0.8 max: 0.8 mean_among_movers: 0.7 agent_readiness_lift: median: 2.1 p75: 2.2 p90: 2.2 max: 2.6 mean_among_movers: 2.2 facet_lift_median_among_movers: discoverability: 7.4 composite_band_moves: thin -> developing: 164 emerging -> thin: 49 developing -> strong: 49 strong -> exemplar: 13 agent_readiness_band_moves: agent-aware -> agent-ready: 639 agent-ready -> agent-native: 68 method: >- each provider's own kin/checks file, the vendor's maps at their stated credit, the scorer's composite formula; from -> to, nothing written