Your AI Assistant, Reading Our Documentation
If you use an AI assistant while you code, you can now connect it directly to our documentation. Instead of guessing from what it learned in training, it can search the current help for ABCpdf, ImageGlue, WordGlue and ABCocr, and read the relevant pages and examples before it answers.
How It Works
Assistants connect to outside information through the Model Context Protocol (MCP), an open standard supported by Claude, ChatGPT, GitHub Copilot in Visual Studio and VS Code, Cursor, Windsurf and many other tools.
Our MCP server gives your assistant a set of tools for searching and reading our documentation. You ask your question as normal and the assistant decides when to look things up.
The Address
Everything you need is this one address:
https://www.websupergoo.com/mcp
It is free, with no sign-in and no API key.
It uses MCP's standard Streamable HTTP transport, so there is nothing to install.
What It Covers
The server covers the documentation for ABCpdf .NET, ImageGlue .NET, WordGlue .NET and ABCocr .NET, including guides, the full API reference and example code. It also covers our product, purchasing and support pages, so your assistant can answer questions about licensing and ordering too.
It is always current. The index is rebuilt automatically as the documentation changes.
Connecting Your Assistant
Most assistants need just a name and the address above. Pick yours below. These tools change often, so if a menu has moved, look for the option to add a custom or remote MCP server.
Claude
For claude.ai, Claude Desktop and the Claude mobile apps:
- Open Customize > Connectors (on some layouts Settings > Connectors).
- Click + and choose Add custom connector.
- Enter the name WebSupergoo and the URL https://www.websupergoo.com/mcp, then click Add. There is no sign-in step.
A connector added on claude.ai is also available in Claude Desktop and the mobile apps. Free plans allow one custom connector. On Team and Enterprise plans an organization owner adds the connector first, under Organization settings > Connectors, and members then connect to it.
Claude Code
Run this once from a command prompt:
claude mcp add --transport http websupergoo https://www.websupergoo.com/mcpAdd --scope user to make it available in all your projects, or --scope project to share it with your team through the project's .mcp.json file. Type /mcp inside Claude Code to check it is connected.
Visual Studio
For GitHub Copilot in Visual Studio 2026, or Visual Studio 2022 version 17.14 or later. Create a file called .mcp.json in your user profile folder (%USERPROFILE%\.mcp.json) to use it with every solution, or in your solution folder to share it through source control:
{
"servers": {
"websupergoo": {
"type": "http",
"url": "https://www.websupergoo.com/mcp"
}
}
}Then open GitHub Copilot Chat, switch to Agent mode and check that the WebSupergoo tools are selected in the tools list.
VS Code
Or open the Command Palette, run MCP: Add Server, choose HTTP and enter the address. Alternatively add this to .vscode/mcp.json in your project:
{
"servers": {
"websupergoo": {
"type": "http",
"url": "https://www.websupergoo.com/mcp"
}
}
}Use Copilot Chat in Agent mode to call the tools.
Cursor
Open Settings > Tools & MCP and choose New MCP Server, or add this to .cursor/mcp.json in your project (or ~/.cursor/mcp.json for all projects):
{
"mcpServers": {
"websupergoo": {
"url": "https://www.websupergoo.com/mcp"
}
}
}Windsurf
Add this to ~/.codeium/windsurf/mcp_config.json. Note that Windsurf uses serverUrl rather than url:
{
"mcpServers": {
"websupergoo": {
"serverUrl": "https://www.websupergoo.com/mcp"
}
}
}ChatGPT
On ChatGPT on the web, open Plugins, click +, choose Add custom MCP server and enter the address. Custom MCP servers are only available on some ChatGPT plans, and OpenAI changes this often, so check OpenAI's help if you don't see the option.
Other Assistants
Any assistant that supports remote MCP servers over Streamable HTTP will work. Give it the address above; no headers, keys or authentication are needed.
What Your Assistant Can Do
Your assistant decides when to use these tools. You just ask your question, and it searches, reads and cross-checks the documentation for you.
When you ask a question, your assistant searches first, using a few key words or the name of a class or method. The search returns the best-matching sections of the documentation, each with a short snippet, and the assistant then reads the most promising pages in full before it answers.
If you paste a line of your own code or an exception message, the assistant can use that too. The search understands API names such as Doc.AddImageUrl or XHtmlOptions, and can pick out sample code when you want an example to adapt.
Searching and Reading
| search | Searches guides, API reference, examples, and purchasing and support pages. Returns the best-matching sections with a snippet of each. |
| fetch | Reads a page in full, as Markdown. |
| lookup_api | Finds a class or member by name, even when misspelt. Also lists enumerations and their values. |
Because the assistant reads the actual reference pages, it can check exact signatures, default values and enumeration members rather than relying on memory. This is where assistants most often go wrong without help.
Finding the Way Around
| contents | Shows the pages around a given page, so the assistant can find related topics. |
| list_products | Lists the products covered and where each one's documentation starts. |
Documentation is easier to understand in context. A property makes more sense alongside the other properties of its class, and a guide often has a companion page of examples. These tools let the assistant explore around a page, just as you would by browsing the contents.
Making the Documentation Better
| report_issue | Flags a problem the assistant found in the documentation, such as a missing page or a wrong example, so it gets fixed for you and everyone else. |
Assistants are surprisingly good at spotting where documentation falls short: a page that was hard to find, an explanation that left a gap, or an example that no longer matches the API. When that happens, once it has answered your question, the assistant can flag the problem directly. It gets put right, so the next answer, whether yours or another developer's, is a better one.
Getting the Best Results
A few small habits make a big difference to the answers you get.
Name the Product
"Using ABCpdf, how do I add a watermark behind the page content?" gets better results than the same question without the product name.
Make It the Default
Most assistants read a project instructions file, such as CLAUDE.md, .github/copilot-instructions.md or AGENTS.md. Adding a line like "For questions about ABCpdf, use the websupergoo MCP server to check the documentation" means you don't have to ask each time.
Be Specific
The more your assistant knows about what you are doing, the better it can search. Include the exact error message, the method or property you are using, and what you expected to happen. A short snippet of the code that is not behaving is often the quickest way to the right page.
Ask for Sources
Ask your assistant to link to the pages it used. You can then read the full explanation yourself, and it is easy to see when an answer has come from the documentation rather than from the assistant's memory.
Mention Your Version
The documentation describes the current release. If you are using an older version, say so, so that your assistant can point out anything that may differ. It is also a good moment to consider whether upgrading would be simpler.
Turn It On in Chat
In some assistants, such as Claude and ChatGPT, you may also need to enable the connector for a conversation from the tools or + menu.
Still Stuck?
Sometimes the answer just is not in the documentation yet. When that happens, your assistant flags the gap, so the answer is there the next time you, or anyone else, need it.
In the meantime, our support page and support team are here for anything the documentation does not answer.
Better Answers, Together
Every question asked through the server makes the documentation better, for you and for every developer who comes after you.
Better Documentation for Everyone
What people search for shows which pages matter most and where things could be explained better. So the documentation keeps improving in exactly the places developers like you need it.
When your assistant finds something wrong or missing in the documentation, it can flag it, much as a helpful colleague might point out a typo. Every report is read and acted on, so the fix is there the next time you, or anyone else, need it. It is a small community effort that makes the documentation better for everyone.
Searches are usually just a few key words, and the server asks assistants to keep your personal details and private code out of their reports. See our privacy page for more.