d365fo-client v0.3.5: Intelligent Cross-Company Queries for D365 F&O

I'm excited to announce d365fo-client v0.3.5, bringing intelligent cross-company query support to Microsoft Dynamics 365 Finance & Operations integration!

This Python package provides a comprehensive toolkit for D365 F&O: OData Client Library, CLI Application, and MCP Server (Model Context Protocol) for AI assistant integration - all with the new smart cross-company capabilities that make working with multi-company data effortless.

The Problem with Cross-Company Queries

When querying data across companies in D365 F&O, you traditionally need to:

Filter by dataAreaId to specify which company's data you want
Remember to add cross-company=true parameter to your OData query
Handle it differently for composite keys

Forgetting step 2 is incredibly common and leads to empty results or cryptic errors. It's one of those things that catches even experienced developers off guard.

Automatic Cross-Company Detection

Version 0.3.5 introduces automatic cross-company detection that eliminates this entire class of errors. The client now analyzes your query filters and automatically enables cross-company mode when needed.

Before (Manual Configuration):

# You had to remember to add cross-company parameter
records = await client.query_entities(
    "SalesOrderHeaders",
    filter="dataAreaId eq 'USMF'",
    cross_company=True  # ❌ Easy to forget!
)

After (Automatic Detection):

# Client detects dataAreaId and enables cross-company automatically
records = await client.query_entities(
    "SalesOrderHeaders",
    filter="dataAreaId eq 'USMF'"
    # ✅ cross-company=true added automatically!
)

How It Works

The query builder analyzes your filter expressions:

Detects dataAreaId presence in filters
Automatically appends cross-company=true to the OData query
Handles both simple and composite keys intelligently
Works with all CRUD operations (query, get, create, update, delete)

This enhancement works across all three interfaces:

Python API: Direct programmatic access
CLI Application: Terminal commands
MCP Server: AI assistant integration (Claude, ChatGPT with plugins, etc.)

Perfect for AI Integration

The MCP (Model Context Protocol) Server support means AI assistants can now intelligently query multi-company D365 F&O data without special configuration or prompting. The cross-company detection works automatically through the MCP protocol!

🤖 AI Integration Bonus: AI assistants using the MCP server can query multi-company data naturally without needing special instructions about cross-company flags. They just filter by dataAreaId and everything works!

What Else is New in 0.3.5

Added:

🏢 Smart Cross-Company Support: Auto-detection for dataAreaId queries (works across Python API, CLI, and MCP server!)
🔑 Composite Key Handling: Cross-company support for complex keys
✨ Zero Configuration: Works out of the box in all interfaces

Improved:

🧪 Test Suite: Removed 618 lines of redundant tests
🎯 Code Quality: Comprehensive formatting and organization
📊 Integration Tests: Enhanced fixtures and consistency

Cleaned Up:

🧹 Removed ~7,000 lines of deprecated MCP server code
🗑️ 15 obsolete test files deleted
📚 Updated documentation to reflect FastMCP architecture

Fixed:

⚠️ All test warnings resolved (clean pytest output!)
🔄 Async handling improved throughout test suite

Updated Dependencies:

📦 13 package updates including MCP SDK 1.21.1, pytest 9.0.1, ruff 0.14.5

Why This Matters

Working with multi-company D365 F&O data just got easier - whether you're using the Python API, CLI commands, or AI assistants via MCP! No more:

❌ Forgetting to set cross-company flags
❌ Manual parameter management
❌ Query errors due to missing configuration

The client is smart enough to understand your intent and do the right thing automatically, across all interfaces.

Get Started

Install the latest version:

pip install d365fo-client==0.3.5
# or with uv
uv add d365fo-client==0.3.5

Links & Resources

📚 Documentation: GitHub Repository
📝 Full Changelog: CHANGELOG.md
📦 PyPI: d365fo-client

Thank You!

Special thanks to the dependabot for keeping our dependencies fresh and the community for continued support!

Have you worked with D365 F&O multi-company scenarios or AI-powered ERP integrations? I'd love to hear your thoughts and experiences in the comments below!