Overview
What you’ll accomplish:- Add the ABBYY Vantage MCP server to GitHub Copilot in VS Code
- Verify the connection is working
- Use the
SearchAbbyyVantageDocumentationtool to search docs from within Copilot Chat
What Is the MCP Server?
The Model Context Protocol (MCP) is an open standard that lets AI assistants connect to external tools and data sources. By adding an MCP server to GitHub Copilot, you give the AI direct access to structured data or capabilities — without copying and pasting content manually. The ABBYY Vantage MCP server exposes a single tool,SearchAbbyyVantageDocumentation, that performs a semantic search across the full Vantage documentation. When Copilot has this server connected, it can look up API parameters, skill configuration options, troubleshooting steps, and more — automatically, as you work.
The MCP server provides read-only access to public documentation. It does not connect to your Vantage tenant, process documents, or require any Vantage credentials.
Prerequisites
- VS Code version 1.99 or later (MCP support required)
- GitHub Copilot extension installed and signed in
No ABBYY Vantage tenant or account is required to use the MCP server.
Add via Command Palette (Recommended)
Use the VS Code Command Palette to register the MCP server:
- Open the Command Palette with Cmd+Shift+P (macOS) or Ctrl+Shift+P (Windows/Linux)
- Type “MCP: Add Server…” and select it
- Select HTTP as the transport type
- Enter the URL:
https://docs.abbyy.com/mcp - Enter the name:
abbyy-vantage - Choose the scope: Workspace or User
| Scope | Config location | When to use |
|---|---|---|
| Workspace | .vscode/mcp.json | One specific project |
| User | User settings via Command Palette | All projects on this machine |
Alternative: Manual Configuration
If you prefer to configure manually, add the server entry to the appropriate configuration file.
Workspace (.vscode/mcp.json)
Workspace (.vscode/mcp.json)
Create or edit
.vscode/mcp.json in your project root:User-level settings
User-level settings
Open the Command Palette (Cmd+Shift+P / Ctrl+Shift+P), type “MCP: Open User Configuration”, and add the following entry:
Verify the Connection
- Open the Command Palette and run “MCP: List Servers” — confirm
abbyy-vantageshows as connected - Open Copilot Chat (click the Copilot icon in the sidebar or press Ctrl+Shift+I / Cmd+Shift+I)
- Switch to Agent mode using the mode selector at the top of the chat panel
- Click the tools icon in the chat input — confirm
SearchAbbyyVantageDocumentationappears in the list - Test with a question:
“What parameters does the Vantage Processing API require to submit a document?”Copilot will invoke
SearchAbbyyVantageDocumentation and return results from the documentation.Available Tools
The ABBYY Vantage MCP server exposes one tool:SearchAbbyyVantageDocumentation
Performs a semantic search across the ABBYY Vantage documentation and returns the most relevant content. Copilot invokes this tool automatically in Agent mode when you ask questions related to Vantage.
Example queries:
- “How do I configure a Document Skill in Vantage?”
- “What does the
minConfidenceparameter do?” - “How do I set up an LLM connection in Vantage?”
- “What export formats does Vantage support?”
- “How do I authenticate with the Vantage REST API?”
Use Cases
API Integrations
Look up endpoint parameters, authentication methods, and request/response formats without leaving your IDE
Skill Configuration
Quickly find configuration options and best practices for Document Skills, Process Skills, and Advanced Designer
Troubleshooting
Search for error messages and known issues to resolve problems faster while debugging integrations
Exploring Capabilities
Discover what Vantage supports — extraction models, supported formats, tenant settings — as you plan your project
Tips for Effective Queries
- Be specific: Include the feature name or API endpoint in your question rather than asking generally
- Use Vantage terminology: Terms like “Process Skill”, “Document Skill”, “Skill Designer”, and “tenant” will return more precise results
- Ask follow-up questions: Copilot retains context, so you can narrow down results with follow-up prompts
- Combine with your code: Paste an API response or config snippet and ask Copilot to explain it using the Vantage docs
- Use Agent mode: MCP tools are only available in Agent mode — make sure the mode selector is set to “Agent” before asking
Removing the MCP Server
You can remove the server using either method: Via Command Palette:- Open the Command Palette (Cmd+Shift+P / Ctrl+Shift+P)
- Run “MCP: List Servers”
- Select
abbyy-vantage - Choose Remove
abbyy-vantage entry from .vscode/mcp.json (workspace) or your user MCP configuration.
Troubleshooting
Server doesn’t appear in “MCP: List Servers” Verify the configuration file was saved correctly and that the JSON is valid (no trailing commas). If you edited the file manually, try reloading VS Code (Developer: Reload Window from the Command Palette). VS Code version too old MCP support requires VS Code 1.99 or later. Check your version via Help → About and update if needed. Tool is not invoked automatically Copilot only invokes MCP tools in Agent mode. Make sure you’ve switched from “Ask” or “Edit” mode to “Agent” using the mode selector at the top of the Copilot Chat panel. If the tool still isn’t being used, try asking your question more explicitly: “Search the ABBYY Vantage docs for…” or mention a specific Vantage feature by name. Search returns irrelevant results Rephrase your query using terms from the Vantage interface or documentation. Avoid generic terms like “workflow” or “settings” on their own — pair them with “Vantage” or a specific feature name.Summary
You’ve successfully:- ✅ Added the ABBYY Vantage MCP server to GitHub Copilot in VS Code
- ✅ Verified the connection using “MCP: List Servers”
- ✅ Learned how to trigger documentation search from Copilot Chat in Agent mode
Next Steps
API Introduction
Get started with the Vantage Processing REST API
Processing Documents
Create a Process Skill and automate document processing via SFTP
Configuring LLM Connections
Connect Vantage to OpenAI or Microsoft Foundry
Prompt-Based Extraction
Use natural language prompts to extract data from documents
MCP Server for Claude Code
Add the same MCP server to Claude Code
