AI Assistance Tools

Pawtograder includes a Model Context Protocol (MCP) server that enables AI assistants to help instructors and TAs support students who are struggling with errors in their programming assignments.

Overview

The MCP server provides AI assistants with secure, read-only access to help request data, discussion threads, student submissions, test results, and build output. This allows AI tools to provide context-aware assistance when helping students debug their code.

Authentication

The MCP server uses Supabase OAuth authentication and is restricted to instructors and graders only. Students cannot access the MCP server.

Privacy Protection

The MCP server is designed with privacy in mind:

Never exposes the users table
Does not expose is_private_profile flags
Provides only the data necessary to help students with their assignments

Available Tools

The MCP server provides the following tools for AI assistants:

Fetch help requests: View active and past help requests from students
Fetch discussion threads: Access discussion board posts and replies
Fetch submissions: Retrieve student submission details
Fetch test results: View autograder test output and results
Fetch build output: Access compilation and build logs

Assignment Context

To provide better context to AI assistants, assignments now include a handout_url field. This field can be set to point to the assignment handout or specification, allowing AI assistants to understand the assignment requirements when helping students.

Using AI Assistance

AI Help Button

When viewing a help request chat or discussion thread, instructors and graders will see an AI Help button. Clicking this button generates an MCP context that can be used with any MCP-compatible AI client.

MCP-Compatible Clients

The MCP server works with any MCP-compatible client, including:

Claude Desktop
Other AI assistants that support the Model Context Protocol

Local Development

When running Pawtograder locally with Supabase, the MCP server is available at:

http://127.0.0.1:54321/mcp

This URL is displayed in the Supabase startup output under “Development Tools”. title: “AI Assistance for Office Hours” description: “Use AI tools to help diagnose student issues and provide better support during office hours” icon: “robot”

description: “Use AI tools to help support students with programming errors and debugging” icon: “sparkles”

AI Assistance for Office Hours

Pawtograder includes AI assistance tools that enable instructors and TAs to use AI assistants (like Claude) to help support students who are struggling with errors in their programming assignments.

Overview

The AI assistance feature uses the Model Context Protocol (MCP) to provide AI assistants with secure, read-only access to relevant course data. This allows AI tools to:

Fetch help requests and discussion threads
Review student submissions and test results
Analyze build output and error messages
Provide context-aware suggestions for helping students

AI assistance is only available to instructors and graders. Students cannot access these tools directly. All access is authenticated through Supabase OAuth.

Features

AI Help Button

When viewing a help request chat or discussion thread, instructors and TAs will see an AI Help button. Clicking this button:

Generates an MCP context for the current conversation
Provides a connection string that can be used with any MCP-compatible client
Includes relevant assignment context, submission data, and test results

Available Tools

The MCP server provides several tools for AI assistants:

Fetch help requests: Retrieve details about student help requests
View discussion threads: Access course discussion posts and replies
Review submissions: Examine student code submissions
Analyze test results: Review autograder output and test failures
Check build output: Inspect compilation errors and build logs

Privacy Protection

The AI assistance system is designed with privacy in mind:

Never exposes the users table or private profile information
Only provides access to course-related data
Requires instructor or grader authentication
All access is logged and auditable

Setting Up AI Assistance

Adding Assignment Context

To help AI assistants provide better support, you can add a handout_url to your assignments. This URL should point to the assignment handout or instructions, which provides important context for the AI assistant. To add a handout URL:

Navigate to the assignment settings
Add the URL to the assignment handout in the handout_url field
Save the assignment

The AI assistant will use this context when analyzing student submissions and providing suggestions.

Using AI Assistance

With Claude Desktop

Click the AI Help button in a help request or discussion thread
Copy the MCP connection string
Add the connection to your Claude Desktop MCP configuration
Claude will have access to the relevant course context

With Other MCP Clients

Any MCP-compatible client can connect to the Pawtograder MCP server using the connection string provided by the AI Help button.

Best Practices

Review AI suggestions: Always review and verify AI-generated suggestions before sharing with students
Use for context: AI assistance is most helpful for understanding error patterns and suggesting debugging approaches
Maintain privacy: Don’t share AI-generated content that includes sensitive student information
Provide feedback: If AI suggestions are unhelpful, use that as an opportunity to improve assignment documentation

Technical Details

The MCP server is built into Pawtograder and uses:

Supabase OAuth for authentication
Read-only database access
Secure token-based authorization
Standard MCP protocol for tool invocation

For more technical details about the MCP implementation, see the Developers documentation. Pawtograder provides AI assistance tools to help instructors and graders provide better support to students. These tools use the Model Context Protocol (MCP) to give AI assistants structured access to student help requests, submissions, and test results.

Getting AI Help

When viewing a help request or discussion thread, click the AI Help button to copy a formatted prompt that includes:

The student’s question and conversation history
Linked assignment details and handout URL
Submission files and test results
Suggested MCP tools for fetching additional context

You can paste this prompt into Claude Desktop, ChatGPT, or any MCP-compatible AI assistant to get intelligent analysis and suggested responses.

Setting Up MCP Access

To use AI assistance with Claude Desktop:

1. Install Claude Desktop

Download from claude.ai/download

2. Create an API Token

Go to Settings → API Tokens in Pawtograder
Click Create New Token
Give it a name (e.g., “Claude Desktop”)
Copy the token (it will only be shown once)

3. Configure Claude Desktop

Add this configuration to your Claude Desktop config file: macOS: ~/Library/Application Support/Claude/claude_desktop_config.json Windows: %APPDATA%\Claude\claude_desktop_config.json

{
  "mcpServers": {
    "pawtograder": {
      "command": "npx",
      "args": ["-y", "@anthropic-ai/mcp-proxy@latest", "YOUR_PAWTOGRADER_URL/functions/v1/mcp-server"],
      "env": {
        "MCP_HEADERS": "{\"Authorization\": \"Bearer YOUR_TOKEN_HERE\"}"
      }
    }
  }
}

Replace YOUR_TOKEN_HERE with the token you created in step 2.

4. Restart Claude Desktop

Quit and reopen Claude Desktop. You should see “pawtograder” listed as an available MCP server.

Available AI Tools

The MCP server provides these tools for AI assistants:

Context Tools

get_help_request - Fetch help request with full conversation history
get_discussion_thread - Fetch discussion thread with replies
get_assignment - Get assignment spec, handout URL, and rubric

Submission Tools

get_submission - Get submission metadata and summary
list_submission_files - See what files were submitted
get_submission_files - Fetch specific source files (supports glob patterns like *.java)

Test Tools

list_submission_tests - See all test results (pass/fail, scores)
get_test_output - Get detailed output for specific tests
get_submission_build_output - Get compilation/build errors

Search Tools

search_help_requests - Find related help requests
search_discussion_threads - Find similar questions from other students

AI Analysis Features

AI assistance is available in multiple contexts:

Help requests: Click AI Help in the help request chat
Discussion threads: Click the AI Help icon in thread actions
Test failures: Click AI Help on failing test results
Build errors: Click AI Help on build error messages
Test insights: Click AI Analyze on common error patterns

Providing Feedback

After using AI assistance, you’ll be prompted to provide feedback on whether the analysis was helpful. This feedback helps improve the AI assistance feature for your course.

AI assistance is only available to instructors and graders. Students cannot access these tools.

Privacy and Security

The MCP server is designed with privacy in mind:

Never exposes student email addresses or user table data
Only accessible to instructors and graders
Requires authentication via API tokens
All data access respects Pawtograder’s existing permissions

Pawtograder includes AI assistance tools that help instructors and TAs support students who are struggling with errors in their programming assignments. The AI Help feature provides context-aware assistance by analyzing student submissions, test results, and discussion threads.

How It Works

The AI assistance system uses the Model Context Protocol (MCP) to enable AI assistants to access relevant student data securely. When helping a student, you can click the AI Help button to generate context that can be shared with MCP-compatible AI clients like Claude.

Available Context

The AI assistance tools can access:

Help requests: Current and past help requests with student descriptions
Discussion threads: Related discussion posts and replies
Submissions: Student code submissions for assignments
Test results: Autograder output and test failures
Build output: Compilation errors and build logs
Assignment handouts: Assignment instructions and requirements (when configured)

Using AI Help

The AI Help button appears in two locations:

Help request chat view: When actively helping a student through office hours
Discussion thread view: When viewing student questions on the discussion board

Click the button to generate an MCP context file that includes all relevant information about the student’s issue. This context can be used with any MCP-compatible AI assistant to get suggestions for helping the student.

Privacy and Security

The AI assistance system is designed with privacy in mind:

Instructor/grader access only: Only users with instructor or grader roles can access AI assistance tools
Supabase OAuth authentication: Secure authentication ensures only authorized users can access student data
No user table exposure: The system never exposes the users table or private profile information
Audit trail: All AI assistance requests are logged for compliance and review

Configuration

To enable AI assistance for an assignment, instructors can add a handout_url field to the assignment. This URL points to the assignment instructions or handout, providing the AI assistant with context about what students are trying to accomplish.

The AI assistance feature requires an MCP-compatible client to use the generated context. The system generates the context but does not directly integrate with any specific AI service.

Best Practices

When using AI assistance to help students:

Review AI suggestions: Always review AI-generated responses before sharing with students
Maintain pedagogical goals: Ensure AI assistance supports learning rather than providing complete solutions
Protect student privacy: Be mindful of what information you share with AI tools
Use assignment context: Set the handout_url field on assignments to help AI understand requirements

Security Considerations

The MCP server requires authentication and is restricted to course staff
All access is logged and auditable
The server provides read-only access to course data
Sensitive user information is never exposed

Review AI suggestions: Always review and verify AI-generated suggestions before sharing them with students
Use as a starting point: Treat AI suggestions as a starting point for your own analysis, not a complete solution
Maintain pedagogical goals: Ensure that AI-assisted help still supports learning objectives and doesn’t simply provide answers
Protect student privacy: Be mindful of student data when using external AI services

MCP Server Setup

For advanced users who want to set up their own MCP server integration, the system provides:

Supabase OAuth authentication flow
RESTful API endpoints for fetching student data
Tools for retrieving help requests, submissions, test results, and build output
Secure token-based access control

Contact your system administrator for details on setting up custom MCP integrations.

Getting Started

Assignments

Polls

Office Hours

Course Features

Automation

​AI Assistance Tools

​Overview

​Authentication

​Privacy Protection

​Available Tools

​Assignment Context

​Using AI Assistance

​AI Help Button

​MCP-Compatible Clients

​Local Development

​description: “Use AI tools to help support students with programming errors and debugging” icon: “sparkles”

​AI Assistance for Office Hours

​Overview

​Features

​AI Help Button

​Available Tools

​Privacy Protection

​Setting Up AI Assistance

​Adding Assignment Context

​Using AI Assistance

​With Claude Desktop

​With Other MCP Clients

​Best Practices

​Technical Details

​Getting AI Help

​Setting Up MCP Access

​1. Install Claude Desktop

​2. Create an API Token

​3. Configure Claude Desktop

​4. Restart Claude Desktop

​Available AI Tools

​Context Tools

​Submission Tools

​Test Tools

​Search Tools

​AI Analysis Features

​Providing Feedback

​Privacy and Security

​How It Works

​Available Context

​Using AI Help

​Privacy and Security

​Configuration

​Best Practices

​Security Considerations

​MCP Server Setup

AI Assistance Tools

Overview

Authentication

Privacy Protection

Available Tools

Assignment Context

Using AI Assistance

AI Help Button

MCP-Compatible Clients

Local Development

description: “Use AI tools to help support students with programming errors and debugging” icon: “sparkles”

AI Assistance for Office Hours

Overview

Features

AI Help Button

Available Tools

Privacy Protection

Setting Up AI Assistance

Adding Assignment Context

Using AI Assistance

With Claude Desktop

With Other MCP Clients

Best Practices

Technical Details

Getting AI Help

Setting Up MCP Access

1. Install Claude Desktop

2. Create an API Token

3. Configure Claude Desktop

4. Restart Claude Desktop

Available AI Tools

Context Tools

Submission Tools

Test Tools

Search Tools

AI Analysis Features

Providing Feedback

Privacy and Security

How It Works

Available Context

Using AI Help

Privacy and Security

Configuration

Best Practices

Security Considerations

MCP Server Setup