uipath-ai-skills

AI skills that turn natural language into production-quality UiPath Studio projects - powered by 94 deterministic Python generators, 71 lint rules, and real Studio-exported templates. No hallucinated XAML. No guessed selectors. Every .xaml file opens cleanly in UiPath Studio.

The problem

LLMs are bad at generating UiPath code. The XAML they produce is typically invalid - wrong namespaces, hallucinated activity names, broken selectors, missing ViewState blocks, incompatible NuGet versions. Most of it won't even open in Studio.

uipath-ai-skills fixes this by constraining and shaping LLM output through deterministic generators, real Studio-exported templates, and a validation layer that catches hallucination patterns before they reach your project.

---

Some examples of what you can build

Automation type	Description	Recommended project
Web form filling	Browser-based data entry with real selectors from Playwright inspection	Sequence or REFramework
Queue dispatching	Read data sources (Excel, DB, API) and populate Orchestrator queues	REFramework Dispatcher
Queue processing	Process queue items with retry logic, exception handling, and status tracking	REFramework Performer
API integrations	HTTP requests with OAuth, retry policies, JSON parsing, credential management	Sequence
Excel processing	Read, filter, transform, and write Excel data with LINQ expressions	Sequence
PDF extraction	Extract text from PDFs using native parsing or OCR	Sequence
Email automation	IMAP retrieval, attachment processing, and sending via Integration Service	Sequence
Desktop automation	Win32/WPF/WinForms app automation with PowerShell-inspected selectors	Sequence or REFramework
Multi-app orchestration	Processes spanning multiple web and desktop applications	REFramework Performer

---

How it works

flowchart LR
    A["Describe automation\nin plain English"] --> B["Agent reads\nSKILL.md routing table"]
    B --> C{"Project\nexists?"}
    C -->|No| D["scaffold_project.py\n(sequence / dispatcher / performer)"]
    C -->|Yes| U{"UI\nautomation?"}
    D --> U
    U -->|Yes| E["Inspect target app(s) UI"]
    E --> F["Write JSON specs\nto disk"]
    U -->|No| F
    F --> G["generate_workflow.py\n(94 deterministic generators)"]
    G --> H["validate_xaml\n(71 lint rules)"]
    H -->|Errors| G
    H -->|Pass| I["generate_object_repository.py\n(selectors -> .objects/)"]
    I --> K["modify_framework.py\n+ resolve_nuget.py"]
    K --> J["Complete UiPath\nproject"]

The skill uses four layers to ensure valid output:

1. Deterministic generators - 94 Python functions that produce structurally correct XAML for each activity type. The LLM writes a JSON spec (which LLMs handle well); generators handle the complex XML structure (which LLMs handle poorly). Every enum value, child element, namespace, and attribute is locked down.

2. Studio-exported templates - every generator is anchored to real XAML exported from UiPath Studio 24.10. Not documentation examples - actual working files.

3. Lint validation - 71 rules that specifically target the hallucination patterns LLMs produce when generating UiPath XAML. Catches issues before you open Studio.

4. Framework wiring - modify_framework.py wires InvokeWorkflowFile calls into framework files, injects variables, replaces scaffold markers, and chains UiElement arguments. generate_object_repository.py builds the .objects/ tree from captured selectors. Both ensure the generated project is structurally complete.

---

Built-in best practices

The skill doesn't just generate valid XAML - it enforces the patterns that experienced UiPath developers apply manually: