Module 1: Supercharge Your IDE with MCP

Module Summary

Goal: Create a custom GitHub MCP Server with Agentic RAG optimization
Estimated Time: 15-20 minutes
Prerequisites: GitHub account, Cursor IDE, xpander.ai account

🚀 In this module, you’ll learn how to build a powerful GitHub search agent with optimized Agentic RAG capabilities and expose it as a Model Context Protocol (MCP) Server to supercharge your Cursor IDE experience. By building this agent, you’ll understand how to create practical AI tools that integrate seamlessly with your development workflow, improving coding efficiency and access to real-time code examples.

🔍 Creating Your GitHub Search Agent

Step 1: Set up xpander.ai

Navigate to the xpander.ai platform:

Terminal Command
# Open in your browser
open https://app.xpander.ai

Sign in with your credentials
In the left navigation menu, go to AI Agents
Click the New AI Agent button to open the Workbench
When prompted by the agent builder, click Skip (we’ll create the agent manually)

Step 2: Configure Agent Settings

In the Builder Workbench, click on the Gear Icon in the top-right corner
In the Agent Builder Settings panel, click on Generate Details

Enter the following prompt to generate your agent description:

Prompt for generating agent
You are GitHub-Search-Agent, an expert at searching and analyzing GitHub repositories, code, users, and topics. You have deep knowledge of GitHub's API and can provide concise, accurate information about code patterns, repository structures, and developer activities.

Now define your agent’s role, goal, and instructions:

Role
You are GitHub-Search-Agent, an expert at searching and analyzing GitHub repositories, code, users, and topics. You have deep knowledge of GitHub's API and can provide concise, accurate information about code patterns, repository structures, and developer activities.

Goal
Your goal is to help users find relevant code, repositories, and developers on GitHub to accelerate their development process. You should prioritize providing the most relevant results and explaining code patterns when found.

Instructions
When searching for code, focus on finding the most relevant implementations
For repositories, highlight the most popular ones with good documentation
When exploring user profiles, focus on their recent contributions and expertise
Prefer repositories with higher star counts and recent activity
For any search query, first try to understand what the user is looking for before running the search
Always provide contextual information about why a result is relevant

Click Save Changes and close the panel

Step 3: Add GitHub Search Tools

Click on the + (plus) button in your agent canvas
Select Apps from the menu
Select GitHub Search Manager
Click Sign in with GitHub Search Manager
Give the interface a name (like “github-search-your-name” or just “github-search”)
Click Save

You should now see the GitHub Search tools available as Agentic Actions in your agent’s canvas.

Step 4: Add GitHub Operations

Add the following GitHub operations to your agent:

Find Code Snippets by Query Terms - GET /search/code
Find Repositories by Criteria - GET /search/repositories
Search Topics by Criteria - GET /search/topics
Find Users by Criteria - GET /search/users
Search Commits by Criteria - GET /search/commits

Your completed agent should look similar to this:

🧪 Testing Your Agent with Raw API Responses

Let’s test your newly created GitHub search agent:

Click on the Tester tab in the xpander.ai interface

Type the following query to search for topics:

Prompt
Search for topics in GitHub related to A2A and MCP

Look at the generated payload by clicking on it. You’ll see something like:

Example AI Generated Payload
{
   "bodyParams": {},
   "queryParams": {
      "q": "A2A MCP",
      "per-page": 10,
      "page": 1
   },
   "pathParams": {}
}

Example response
{
 "total_count": 2,
 "incomplete_results": false,
 "items": [
     {
         "name": "a2a-mcp",
         "created_at": "2025-04-30T00:52:27Z",
         "updated_at": "2025-04-30T00:52:27Z",
         "featured": false,
         "curated": false,
         "score": 1
     },
     {
         "name": "a2a-vs-mcp",
         "created_at": "2025-04-30T00:52:27Z",
         "updated_at": "2025-04-30T00:52:27Z",
         "featured": false,
         "curated": false,
         "score": 1
     }
 ]
}

Now try a repositories search:

Prompt
Find cool repositories related to Agent 2 Agent and MCP

AI Generated payload
{
   "bodyParams": {},
   "queryParams": {
      "q": "Agent 2 Agent MCP",
      "sort": "stars",
      "order": "desc",
      "per-page": 5,
      "page": 1,
      "xp_rag_query": "Agent 2 Agent MCP",
      "xp_rag_top_k": 5
   },
   "pathParams": {}
}

Notice that the API response is huge and not optimal for AI consumption! This is where Agentic RAG comes in to help filter and optimize these responses.

🧠 Optimizing Responses with Agentic RAG

Optimizing Repository Search

Let’s improve the repository search results using Agentic RAG:

Return to the Builder tab
Click on the gear icon next to the Find Repositories by Criteria operation
Click on Advanced and then switch to Raw Editor (this allows the AI to send natural language queries, and the server will perform semantic search on the API response)
Configure the search and return fields:

Searchable Fields (fields the AI can search against):
Searchable Fields Configuration
```
[
  "items[].full_name",
  "items[].description",
  "items[].topics"
]
```
Returnable Fields (data the AI will receive):
Returnable Fields Configuration
```
[
  "items[].html_url",
  "items[].full_name",
  "items[].description"
]
```
Save your changes
Return to the Tester tab and run the repositories query again:
Prompt
```
Find cool repositories related to Agent 2 Agent and MCP
```

Notice how much cleaner and more focused the response is now! Instead of overwhelming JSON data, you get just the meaningful content that matters to your query.

Optimizing User Search

Next, let’s optimize the user search operation:

Return to the Builder tab
Click on the gear icon next to the Find Users by Criteria operation
Go to Advanced → Raw Editor

Configure the fields:

Searchable Fields:

User Search Configuration
[
  "items[].login"
]

Returnable Fields:

User Return Configuration
[
  "items[].html_url",
  "items[].login", 
  "items[].avatar_url"
]

Save your changes

In the Tester tab, run:

Prompt
Find GitHub users related to MCP and Agent 2 Agent protocol

Optimizing Commit Searches

Now, let’s configure commit search operations:

Return to the Builder tab
Click on the gear icon next to the Search Commits by Criteria operation
Go to Advanced → Raw Editor

Configure the fields:

Searchable Fields:

Commit Search Configuration
[
  "items[].commit.message",
  "items[].author.login",
  "items[].commit.author.name",
  "items[].repository.full_name"
]

Returnable Fields:

Commit Return Configuration
[
  "items[].html_url",
  "items[].sha",
  "items[].commit.message",
  "items[].commit.author.date",
  "items[].author.login", 
  "items[].repository.html_url"
]

Save your changes
In the Tester tab, try a follow-up question:
Follow-up Prompt
```
Summarize their recent commits
```

Optimizing Code Snippets Search

Finally, let’s optimize the code snippet operation:

Return to the Builder tab
Click on the gear icon next to the Find Code Snippets by Query Terms operation
Go to Advanced → Raw Editor

Configure the fields:

Searchable Fields:

Searchable Fields Configuration
[
  "items[].name",
  "items[].path",
  "items[].repository.full_name",
  "items[].repository.description"
]

Returnable Fields:

Returnable Fields Configuration
[
  "items[].html_url",
  "items[].path",
  "items[].repository.html_url",
  "items[].repository.full_name"
]

Save your changes

In the Tester tab, run:

Prompt
Can you find the implementation of A2A code snippet?

The optimized results should look similar to this:

Example Output
Here are some code snippets related to "A2A" found on GitHub:

1. **Repository:** [hyunjun/bookmarks](https://github.com/hyunjun/bookmarks)
   - **File:** [c.md](https://github.com/hyunjun/bookmarks/blob/b0dce61583620ed1bff22408dfae28563e3cf28f/c.md)

2. **Repository:** [mwherman2000/did-uri-spec](https://github.com/mwherman2000/did-uri-spec)
   - **File:** [FAQ.md](https://github.com/mwherman2000/did-uri-spec/blob/deff7e273046574ec3703313b1bb833a4d060cbc/FAQ.md)

3. **Repository:** [zeromake/library](https://github.com/zeromake/library)
   - **File:** [TOC.md](https://github.com/zeromake/library/blob/6073134ad593c54897efcb3339f4a0662bc7a7b1/TOC.md)

4. **Repository:** [tansuotv/IPTVindex](https://github.com/tansuotv/IPTVindex)
   - **File:** [tv.txt](https://github.com/tansuotv/IPTVindex/blob/869d6d43ae2316cc4b0628affc9b87650a783c5a/tv.txt)

5. **Repository:** [kitnil/notes](https://github.com/kitnil/notes)
   - **File:** [s3.org](https://github.com/kitnil/notes/blob/da3da9c30cbebee0075fb92f58f8fc3dac7e7e0f/s3.org)

📄 Adding Code Reading Capabilities

Let’s enhance our agent by adding the ability to read code directly from GitHub URLs:

In the left navigation menu, go to Cloud Functions
Click the New button
Add the following code:

GitHub Code Reader Function
# === GitHub Code Reader ===
# Contract-name: xpander_run_action
# Purpose: Retrieve any file's raw text from GitHub (public or private) so downstream
#          agents can parse the code.

# import requests – auto-imported by the runtime
import os

def xpander_run_action(repo_full_name: str, file_path: str, ref: str = "main"):
    # GitHub Code Reader
    # ------------------
    # Args:
    #     repo_full_name : str  | "owner/repo", e.g. "org/project"
    #     file_path      : str  | path within the repo, e.g. "src/app.py"
    #     ref            : str  | branch, tag, or commit SHA (default "main")
    # Optional ENV:
    #     GIT_TOKEN            | Personal Access Token for private repositories.
    #                           If present, it's sent as a Bearer token.
    # Returns:
    #     str – the file's text on success, or an error string on failure.

    raw_url = f"https://raw.githubusercontent.com/{repo_full_name}/{ref}/{file_path}"

    headers = {}
    token = os.getenv("GIT_TOKEN")
    if token:
        headers["Authorization"] = f"Bearer {token}"

    try:
        resp = requests.get(raw_url, headers=headers, timeout=15)
        resp.raise_for_status()
    except requests.RequestException as exc:
        return f"Request failed: {exc}"

    return resp.text

Name it GitHub Code Reader and click Save
Return to your agent in the AI Agents section
Click the + button
Select Custom Action
Add the GitHub Code Reader function to your agent canvas
Click Save -> Deploy.

Run the following prompt

Example Prompt
How can I increase throughput using cross-region inference with Amazon Bedrock and boto3? Download and provide the real python code snippet from AWS repos, not a link

The screenshot is from the agent’s WebUI. You can access yours by clicking the ‘Chat’ button on the canvas — the WebUI URL will appear

🌎 Enhancing Your Agent with Web Search

Your GitHub search agent currently can only access information available on GitHub. Let’s add web search capabilities to make it more versatile:

Return to your agent canvas
Click the + button
Select Built-in Actions
Add Fetch Tavily AI Insights (for comprehensive web searching)
Click on the gear icon in your agent canvas to edit the agent settings

Update your agent’s instructions by adding these important rules:

Updated Instructions
ALWAYS start with Tavily AI Insights to gather general information, but NEVER stop there
After using Tavily, ALWAYS follow up with relevant GitHub searches to find concrete code examples and repositories
For every user query, you must provide both web-based context AND GitHub-specific examples

Save your changes

In the Tester tab, try a query that requires recent information:

Test Prompt
What are the latest changes to the A2A protocol specification? Show me example implementations.

Your agent should now provide more comprehensive answers by combining web search results with GitHub code examples:

Sometimes, our autonomous agent might choose to use GitHub before Tavily, despite our explicit instructions. This happens because the AI model sees all available tools at once and makes decisions based on its understanding of the prompt and tool descriptions.

This reveals a fascinating design challenge in AI systems: How can we create truly autonomous agents capable of making intelligent decisions while ensuring they reliably follow critical business logic when needed? 🤔

🔄 Creating the Agent Dependency Graph

In production, to ensure your agent follows the right search strategy, we should always create a dependency graph that deterministically enforces specific business logic behavior:

In your agent canvas, click on the Graph View button
Set up a workflow that requires an internet search before accessing GitHub operations:
- Connect the Tavily AI Insights node as a prerequisite to the GitHub operations
- Ensure the flow follows a logical progression

Repeat the query with explict request to not start with Tavily

test prompt
What are the latest changes to the A2A protocol specification? Show me example implementations. Use GitHub without Tavily

Validate additonal queries:

Test Prompt
Using only Google's A2A protocol implementation, create an agent that responds with 'hello world'. Provide the Python code.

Verify that the AI follows the correct workflow - first searching for information before accessing GitHub operations:

Congratulations! You’ve built a reliable AI Agent that balances autonomy with guardrails. By creating this structured workflow, you’ve effectively addressed both the API overflow challenge and solved the autonomous multi-step challenges

🌐 Exposing Your Agent as an MCP

The final step is to expose your optimized GitHub agent as a Model Context Protocol (MCP) server:

In your agent canvas, click the + button
Select Source → MCP
Click Deploy

After deployment, you’ll receive a unique URL for your MCP server

Configure this endpoint in your Cursor IDE by editing the settings file:

Create or edit ~/.cursor/mcp.json:

mcp.json
{
  "mcpServers":{
    "xpander-github-mcp": {
      "command": "npx",
      "args": [
        "xpander-mcp-remote",
        "https://mcp.xpander.ai/your-server-url/"
      ]
    }
  }
}

Make sure to replace https://mcp.xpander.ai/your-server-url/ with your actual MCP server URL from the deployment step!

Restart Cursor IDE to apply the changes
Test your MCP integration by asking a question in Cursor:
Prompt to Agent (via MCP)
```
search for code samples of xpander-ai sdk
```

✅ Checkpoint

Congratulations! By completing this module, you should now be able to:

Create and configure a GitHub search agent with optimized Agentic RAG
Customize searchable and returnable fields for each API operation
Add custom code reading capabilities to your agent
Structure agent workflows using dependency graphs
Deploy your agent as an MCP endpoint for Cursor IDE

🔄 Next Steps

Now that you’ve supercharged your IDE with an optimized GitHub MCP Server, you’re ready to build your first coding agent in the next module.

Workshops

​🔍 Creating Your GitHub Search Agent

​Step 1: Set up xpander.ai

​Step 2: Configure Agent Settings

​Step 3: Add GitHub Search Tools

​Step 4: Add GitHub Operations

​🧪 Testing Your Agent with Raw API Responses

​🧠 Optimizing Responses with Agentic RAG

​Optimizing Repository Search

​Optimizing User Search

​Optimizing Commit Searches

​Optimizing Code Snippets Search

​📄 Adding Code Reading Capabilities

​🌎 Enhancing Your Agent with Web Search

​🔄 Creating the Agent Dependency Graph

​🌐 Exposing Your Agent as an MCP

​✅ Checkpoint

​🔄 Next Steps

🔍 Creating Your GitHub Search Agent

Step 1: Set up xpander.ai

Step 2: Configure Agent Settings

Step 3: Add GitHub Search Tools

Step 4: Add GitHub Operations

🧪 Testing Your Agent with Raw API Responses

🧠 Optimizing Responses with Agentic RAG

Optimizing Repository Search

Optimizing User Search

Optimizing Commit Searches

Optimizing Code Snippets Search

📄 Adding Code Reading Capabilities

🌎 Enhancing Your Agent with Web Search

🔄 Creating the Agent Dependency Graph

🌐 Exposing Your Agent as an MCP

✅ Checkpoint

🔄 Next Steps