- Home
- Skills
- Agents & Orchestration
- UML Modeler for Visual Paradigm (MCP skill)
More screenshots
Works with the AI Agents you already use
UML Modeler for Visual Paradigm (MCP skill)
Your agent builds real UML and ER diagrams in Visual Paradigm 18.1 from plain text: sequence, class, use case, activity, state.
$9.99
UML Modeler for Visual Paradigm (MCP skill)
Example session with this skill installed
Draw a sequence diagram in Visual Paradigm for placing an order: the customer submits the order form, the form calls OrderController.placeOrder(cart), the controller validates, creates an Order and adds every item in a loop. If payment succeeds, it marks the order paid, otherwise the form shows an error. Then refer to the "Send notification" interaction, return the order id and destroy the Order.
- Read your context and instructions
- Compiled the uml modeler
Summary
Created the sequence diagram
"Place order" in the open project with one vp_build_sequence call, checked the exported image and saved the project.
| Item | Result |
|---|---|
| Participants | Customer (actor), form : OrderForm (boundary), controller : OrderController (control), PaymentService, order : Order (created during the flow) |
| Messages | 14, with activation bars, self call and reply messages |
| Fragments | loop [for each item], alt [payment succeeded] / [else] |
| Interaction use | ref Send notification |
| Lifecycle | new Order() create message, «destroy» at the end |
| Frame | sd Place order |
Next steps
- Open the diagram in Visual Paradigm and adjust anything by hand: it is a native model, not an image
- Ask for the matching class diagram: the agent reuses OrderForm, OrderController and Order instead of creating duplicates
- Ask for a use case specification for "Place order"
Connects securely to your tools. The creator never sees your data.
About this skill
Requires Visual Paradigm 18.1 or later (the free Community Edition works).
Stop drawing UML by hand. Describe the system in plain words, and your AI agent builds real, editable models in your Visual Paradigm project: not a picture, not PlantUML text, but native VP diagrams you can keep working on.
What your agent can build
- Sequence diagrams with activation bars, self calls, create/destroy, async and reply messages, combined fragments (alt, opt, loop, par, break) with guards, interaction uses (ref) and the
sdframe - Class diagrams: classes, interfaces, enumerations, attributes and operations, inheritance, realization, associations with roles and multiplicities, aggregation/composition, qualifiers, association classes, stereotypes
- Use case diagrams with system boundary, include/extend, actor generalization, plus full use case specifications (pre/post conditions, flows of events)
- Activity diagrams with swimlanes, decisions and guards
- State machine and ER diagrams (tables, columns, keys, relationships)
How the skill works
The skill teaches your agent a careful, reliable workflow for the Visual Paradigm MCP tools (vp_*):
- Checks which project is open and reuses existing elements instead of duplicating them
- Picks the right tool and builds each diagram in one large step
- Lays the diagram out automatically
- Exports a PNG, looks at it, and fixes what is wrong
- Saves on purpose and handles Visual Paradigm dialogs safely
It also includes a reference of worked examples for every diagram type, so the agent gets the arguments right the first time.
Requirements
- Visual Paradigm 18.1 or later (the free Community Edition works)
- The free open-source Visual Paradigm MCP plugin. Install and connection steps for every supported agent are inside the skill.
Works with: Claude Code, Codex CLI, OpenCode, Antigravity CLI (tested). Gemini CLI, Cursor, VS Code / GitHub Copilot and Cline are supported by config (not tested yet).
Independent project, not affiliated with or endorsed by Visual Paradigm International. Visual Paradigm is a trademark of its owner.
Requires Visual Paradigm 18.1 or later and the free Visual Paradigm MCP plugin. Visual Paradigm must be running while the agent works. Tested on Windows 11 with Visual Paradigm CE 18.1; macOS and Linux are expected to work but are not tested.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before any payment.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scan
Security scan passed 1 month ago
Requires
Permissions
No special permissions declared or detected