pbi-cli  by MinaSaad1

AI agent for Power BI semantic models and reports

Created 2 months ago
379 stars

Top 74.9% on SourcePulse

GitHubView on GitHub
Project Summary

Summary

MinaSaad1/pbi-cli offers a command-line interface tool designed to empower AI agents, specifically Claude Code, with advanced capabilities for interacting with Power BI semantic models and reports. It bridges the gap between AI and Power BI development, enabling users to automate tasks like DAX query generation, report construction, model auditing, and custom visual development through natural language prompts. This tool is beneficial for Power BI developers, analysts, and users seeking to leverage AI for increased productivity and streamlined Power BI asset management.

How It Works

The architecture comprises two distinct layers. The Semantic Model layer facilitates direct, in-process interaction with Power BI Desktop using .NET TOM/ADOMD interop from Python, ensuring sub-second execution without external servers or MCP services. The Report layer operates directly on PBIR (Enhanced Report Format) JSON files, enabling read/write operations for .pbip projects without requiring an active Power BI connection. This dual-layer approach allows for comprehensive management of both the data model and the report presentation layer.

Quick Start & Requirements

  • Primary Install: pipx install pbi-cli-tool (recommended for automatic PATH management) or pip install pbi-cli-tool.
  • Initial Setup: Run pbi-cli skills install to register Claude Code skills, followed by pbi connect while Power BI Desktop is open with a .pbix file.
  • Prerequisites: Windows operating system, Python 3.10+, and Power BI Desktop installed.
  • Alternative Install: Clone the repository from https://github.com/MinaSaad1/pbi-cli.git and install from source.
  • Note: Users installing via pip may need to manually add the Scripts directory to their system PATH.

Highlighted Details

  • Provides 13 distinct Power BI skills for AI agents, covering DAX, modeling, deployment, security, diagnostics, and reporting.
  • Enables AI-driven creation of measures in bulk, debugging of DAX queries, snapshotting/restoring semantic models, and auditing for issues.
  • Supports AI-assisted report building, including scaffolding, managing 32 visual types, pages, themes, and filters.
  • Features an AI-powered custom visual authoring workflow, automating TypeScript development, packaging (.pbiviz), and integration into reports.
  • Offers 27 command groups for comprehensive management of semantic models and reports, with --json output for scripting and AI integration.
  • Includes a REPL mode for interactive command-line use with tab completion and command history.

Maintenance & Community

The project is maintained by MinaSaad1. Specific community channels (like Discord/Slack) or details on major contributors/sponsorships are not explicitly detailed in the README.

Licensing & Compatibility

The core pbi-cli-tool package is distributed under the MIT License. However, bundled Microsoft Analysis Services client library DLLs (Microsoft.AnalysisServices.*.dll) are redistributed unmodified under their separate Microsoft Software License Terms. Users must comply with both the MIT license for the tool and the Microsoft terms for the included DLLs, as detailed in THIRD_PARTY_LICENSES.md.

Limitations & Caveats

The tool is primarily designed for Windows environments and requires Power BI Desktop to be running for semantic model operations. The bundled Microsoft DLLs are subject to separate licensing terms that users must review and adhere to.

Health Check
Last Commit

1 hour ago

Responsiveness

Inactive

Pull Requests (30d)
0
Issues (30d)
2
Star History
61 stars in the last 30 days

Explore Similar Projects

Starred by Shizhe Diao Shizhe Diao(Author of LMFlow; Research Scientist at NVIDIA), Michael Chiang Michael Chiang(Cofounder of Ollama), and
7 more.

openbench by groq

0%
782
Provider-agnostic LLM evaluation infrastructure
Created 10 months ago
Updated 1 month ago
Feedback? Help us improve.