MLCopilot - MarkLogic AI Assistant

Your intelligent companion for MarkLogic development

MLCopilot brings AI-powered assistance to MarkLogic developers working with XQuery and XSLT, combining syntax highlighting, static analysis, intelligent code generation, and task management for MLCP and CoRB workflows.

Features

XQuery & XSLT Language Support

• Full XQuery syntax highlighting for .xq, .xqy, .xquery, .xql, .xqm files
• Full XSLT syntax highlighting for .xsl, .xslt files
• Theme-aware token colors (light and dark) for variables, functions, types, keywords, and operators
• Language Server Protocol (LSP) integration for real-time XQuery static error detection

Code Snippets

• XQuery snippets — version declaration, module namespace, imports, function/variable declarations, FLWOR expressions, and more
• XSLT snippets — stylesheet skeletons (2.0 and 3.0), templates, xsl:for-each, xsl:if, xsl:choose, and more

MarkLogic Function IntelliSense

• Autocomplete — 3000+ MarkLogic built-in function completions with signatures and documentation, triggered by : and -
• Hover documentation — function signature, parameters, return type, and examples on hover

AI-Powered Features

Natural Language to XQuery

• Command: MLCopilot: Natural Language to XQuery
• Key: Ctrl+Shift+X / Cmd+Shift+X
• Describe your intent in plain English; AI generates optimized XQuery code
• Output includes a copy-to-clipboard button in the side panel

Explain Selected XQuery Code

• Command: MLCopilot: Explain Selected XQuery Code
• Key: Ctrl+Shift+E / Cmd+Shift+E (requires selection)
• Context menu: Right-click on selected XQuery → "Explain Selected XQuery Code"
• AI explains selected code with constructs, purpose, and gotchas in a side panel

Hover Error Explanation

• Hover over a syntax error in XQuery to get an AI-generated explanation and fix suggestion

MLCP Task Generator

• Command: MLCopilot: Generate MLCP Task
• Key: Ctrl+Shift+M / Cmd+Shift+M (in build.gradle)
• Context menu: Right-click in build.gradle → "Generate MLCP Task"
• Interactively generates MarkLogic Content Pump (MLCP) Gradle task definitions
• Supported task types: CSV Import, TSV Import, JSON Import, XML Import, Export
• Prompts for task name, database, collection, and generates full configuration with batch size, thread count, and transaction size

CoRB Task Generator

• Command: MLCopilot: Generate CoRB Task
• Key: Ctrl+Shift+C / Cmd+Shift+C (in build.gradle)
• Context menu: Right-click in build.gradle → "Generate CoRB Task"
• Interactively generates Content on Request (CoRB) Gradle task definitions
• Prompts for task name, URIs type (module-based or file-based), thread count, and optional export file

MLCP & CoRB IntelliSense in build.gradle

• MLCP autocomplete — property completions for IMPORT, EXPORT, and COPY commands, triggered by = and space
• MLCP hover — property description, type, valid values, examples, and applicable commands
• CoRB autocomplete — property completions inside CoRB task blocks, with category, type, and examples
• CoRB hover — property documentation on hover

Query Editor

• Command: MLCopilot: Open Query Editor
• Dedicated MLCopilot Query Activity Bar panel for writing and running XQuery directly against a MarkLogic server
• Full webview editor with syntax-aware input and result display
• Complements the Tasks Panel — write ad-hoc queries without leaving VS Code

Tasks Panel

• Sidebar: Activity bar icon opens "MLCopilot Tasks" panel
• Auto-scans workspace build.gradle files to list all MLCP and CoRB tasks
• Shows tasks in organized tables with MLCP/CoRB type badges
• One-click DEV and PROD or LOCAL run buttons that spawn a terminal
• Real-time running status with animation; prevents duplicate executions
• Refresh command: MLCopilot: Refresh Tasks (refresh icon in panel title)

Multi-Provider AI Support

• Command: MLCopilot: Switch AI Provider
• Switch between AI providers at any time via quick pick:
  • Auto — tries Claude Code → ChatGPT → Gemini automatically
  • Claude Code — free, no API key needed (uses the Claude Code VS Code extension)
  • ChatGPT — OpenAI API (requires mlcopilot.openai.apiKey)
  • Gemini — Google Gemini API (free tier available, requires mlcopilot.gemini.apiKey)

Installation

1. Open VS Code
2. Go to Extensions (Ctrl+Shift+X)
3. Search for "MLCopilot"
4. Click Install

Getting Started

1. Open any .xqy or .xsl file — syntax highlighting and IntelliSense work automatically
2. Errors appear in the Problems panel via the XQuery language server
3. For AI features, configure your provider in settings (see below)

Extension Settings

Commands & Keyboard Shortcuts

Requirements

• VS Code 1.115.0 or higher
• (Optional) API key for ChatGPT or Gemini if not using Claude Code

Known Issues

• Static analysis currently supports XQuery 1.0-ml
• XSLT error detection is limited to syntax validation

Roadmap

• [ ] Enhanced XSLT support
• [ ] Code completion for MarkLogic server-side modules
• [ ] Integration with MarkLogic server for live validation
• [ ] Snippet library for common patterns

Contributing

Found a bug or have a feature request? Open an issue on GitHub.

License

MIT License — see LICENSE file for details.

Author

Akbar Ali
Senior MarkLogic Developer | MarkLogic Certified (v10, v12)