solution-decision-framework

Routes Kingdee K/3 Cloud customization requests across standard features, BOS configuration, Python plugins, and DLL plugins.

24|7|Updated Apr 20, 2026
One-click install
npx skills add https://github.com/qiaolei227/opendeploy --skill solution-decision-framework-qiaolei227
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: solution-decision-framework
Source: https://github.com/qiaolei227/opendeploy/tree/main/knowledge/skills/k3cloud/solution-decision-framework
Command: npx skills add https://github.com/qiaolei227/opendeploy --skill solution-decision-framework-qiaolei227

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? ERP implementation agents often jump straight to writing Python plugins for every client request, creating unnecessary technical debt in customer environments. This Skill enforces a four-layer decision tree so each functional requirement lands on the lowest-cost viable layer of Kingdee K/3 Cloud customization. ## Core Features & Use Cases - Four-Layer Decision Tree: Evaluates requirements against standard K/3 Cloud features first, then BOS no-code configuration, then Python plugins, then DLL plugins, stopping at the first layer that solves the problem. - Misjudgment Catalog: Lists common requests that look like custom development but are actually standard features, such as credit control enablement, list filters, and built-in alert platform notifications. - Tool Boundary Awareness: Documents which BOS operations OpenDeploy automates (extension creation, Python plugin registration, convert plugins) and which require manual BOS Designer work (field extensions, approval flows, print templates). - Use Case: A client asks to add a validation rule on sales order save. Instead of immediately creating an extension, the agent first checks standard features and BOS business rules, confirms a Python form plugin is the right layer, then follows the design-approval-register-verify workflow. ## Quick Start Load this skill whenever a new functional requirement arrives for Kingdee K/3 Cloud and walk the requirement through the four-layer decision tree before writing any code.

Frequently Asked Questions about solution-decision-framework

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I decide between BOS configuration and a Python plugin in Kingdee K/3 Cloud?▼

Check standard product features first, then BOS Designer no-code configuration, and only use a Python plugin when the requirement is event-driven logic like BeforeSave validation or field-change cascading. Stop at the first layer that solves the problem.

When should I use a Python plugin instead of standard K/3 Cloud features?▼

Only after confirming the standard feature index and BOS configuration cannot satisfy the requirement. Common cases needing Python plugins include save-time validation, field auto-fill on change, button-triggered actions, and convert-time field mapping.

Can OpenDeploy automate BOS field extensions and approval flows?▼

No. OpenDeploy currently automates BOS extension creation and Python plugin registration, but field extensions, business rules, convert rules, approval flows, and print templates require manual configuration in BOS Designer.

What are the limitations of Python plugins in K/3 Cloud?▼

Python plugins cannot handle cross-document aggregation reports, high-frequency big-data processing, complex threading due to IronPython limits, third-party NuGet dependencies, or print widget customization. Those cases require .NET DLL development.

How do I handle multi-entry convert field mapping in K/3 Cloud?▼

Standard convert rules support only one main associated entity. For cross-entry or child-entity mapping, register a PythonConvertPlugIn and populate link data in the OnAfterCreateLink event with the six required fields including FlowId and STableName.