Horizun PBI MCP
OfficialRelated Servers
Alternatives to Horizun PBI MCP
No user-submitted related servers found.
Related Servers
- AlicenseBqualityBmaintenanceMCP server for interacting with local Power BI Desktop and .pbip projects, enabling DAX queries, model documentation, measure editing, and PBIR visual management via natural language.90534 PyPI1Apache 2.0
- AlicenseNot gradedqualityDmaintenanceMCP server for connecting to Microsoft Fabric and Power BI semantic models, enabling workspace browsing, dataset management, TMSL retrieval, DAX queries, model creation/editing, and includes a Best Practice Analyzer with 71 rules.2MIT
- AlicenseCqualityBmaintenanceLocal-first MCP server for Power BI Desktop automation. Automate semantic model changes, DAX, Power Query, Excel, and report layout from MCP-capable AI clients.1001MIT
- AlicenseNot gradedqualityDmaintenanceLocal MCP server for editing Power BI Project reports saved in PBIR format.GPL 2.0
- FlicenseNot gradedqualityDmaintenanceMCP server for reading and analyzing Fabric semantic models. Supports getting model definitions and executing DAX queries against Power BI datasets.17-
- FlicenseNot gradedqualityDmaintenanceA comprehensive MCP server for Power BI DAX query execution, validation, optimization, and intelligent assistance with tools for schema analysis, template generation, and health monitoring.-
TDQS
Scored across 139 tools
Many tools overlap in purpose despite being individually documented: page creation exists as pbi_create_page_from_spec, pbi_apply_page_spec, pbi_compose_page, pbi_generate_report_page and pbi_build_dashboard; validation is split across pbi_validate_tmdl/pbip/page_spec/measures/generated_page/desktop_render; and layout fixing spans pbi_normalize_page_layout, pbi_normalize_report, pbi_arrange_visuals and pbi_align_visuals. The very long descriptions do explain the distinctions and cross-reference each other, which lifts it above pure ambiguity, but at 139 tools an agent will frequently hesitate between near-synonymous options.
Every tool uses a uniform pbi_ prefix with snake_case verb_noun structure (pbi_create_measure, pbi_delete_visual, pbi_validate_page_spec, pbi_list_tables). Even status/noun-only helpers like pbi_health_check, pbi_capabilities and pbi_session_info fit the same lexical pattern without mixing camelCase or alternate verbs.
139 tools is an extreme mismatch for a workable surface, far past the 50+ threshold, and even the server's own pbi_start_here description admits that '132 tools with good names are still 132 tools.' The domain is broad (model, report, layout, export, SharePoint) but the count is not scoped to what an agent can select among reliably.
Coverage is unusually exhaustive: full CRUD over measures, columns, relationships, hierarchies, pages, visuals and bookmarks, plus lifecycle support (convert, validate, backup, journal, recover, purge, export, finalize) and model/report inspection. Real gaps are hard to identify for a Power BI automation server.