Name: debug-issue
Author: yubo-yue

Prompt file imported from yubo-yue/yubo-python (.github/prompts/debug-issue.prompt.md). Copyright stays with the author.

Debug Python Issues

Systematic debugging assistance for Python code issues, focusing on MCP server development and AI/ML experimentation.

Debugging Process

1. Problem Identification

Gather error messages, stack traces, and symptoms
Identify when the issue occurs (startup, runtime, specific conditions)
Determine affected components (MCP tools, server, dependencies)
Check recent changes that might have introduced the issue

2. Investigation Strategy

Error Analysis: Parse stack traces and error messages
Code Review: Examine relevant code sections for issues
Environment Check: Verify Python version, dependencies, and configuration
Logging Analysis: Review application logs and debug output

3. Common Issue Categories

MCP Server Issues

Tool registration and schema generation problems
Transport configuration and connection issues
Type hint errors affecting schema validation
Async/await usage problems

Python Code Issues

Import errors and dependency conflicts
Type annotation problems
Exception handling and error propagation
Performance bottlenecks and memory issues

Environment Issues

Conda environment configuration problems
Package version conflicts
Missing dependencies or incorrect installations
Path and module resolution issues

Debugging Tools & Techniques

Built-in Debugging

Use print() statements for quick debugging
Leverage Python's pdb debugger for interactive debugging
Add logging with appropriate levels (debug, info, warning, error)
Use traceback module for detailed error information

Testing & Validation

Create minimal reproducible examples
Write unit tests to isolate issues
Use pytest with verbose output
Mock external dependencies to isolate problems

MCP Specific Debugging

Test tools independently before LLM integration
Use uv run mcp dev server.py for development testing
Validate schemas with different parameter types
Check Claude Desktop integration logs

Resolution Approach

Reproduce: Create a minimal case that demonstrates the issue
Isolate: Narrow down to the specific component or function
Analyze: Understand the root cause of the problem
Fix: Implement a targeted solution
Test: Verify the fix works and doesn't break other functionality
Document: Update code comments or documentation if needed

Provide the error message, stack trace, or describe the issue you're experiencing, and I'll help debug it systematically.

Install debug-issue and it becomes one more candidate the router can pick - when it fits.

A listing is a routing target with a manifest: what it does, which domains it covers, what it costs and who publishes it. Once installed it sits beside your own models and tools, is scored like any other candidate for each request, and shows up in the trace when it wins. Ratings come from workspaces that installed it, one per account.

# Debug Python Issues Systematic debugging assistance for Python code issues, focusing on MCP server development and AI/ML experimentation. ## Debugging Process ### 1. Problem Identification - Gather error messages, stack traces, and symptoms - Identify when the issue occurs (startup, runtime, specific conditions) - Determine affected components (MCP tools, server, dependencies) - Check recent changes that might have introduced the issue ### 2. Investigation Strategy - **Error Analysis**: Parse stack traces and error messages - **Code Review**: Examine relevant code sections for issues - **Environment Check**: Verify Python version, dependencies, and configuration - **Logging Analysis**: Review application logs and debug output ### 3. Common Issue Categories #### MCP Server Issues - Tool registration and schema generation problems - Transport configuration and connection issues - Type hint errors affecting schema validation - Async/await usage problems #### Python Code Issues - Import errors and dependency conflicts - Type annotation problems - Exception handling and error propagation - Performance bottlenecks and memory issues #### Environment Issues - Conda environment configuration problems - Package version conflicts - Missing dependencies or incorrect installations - Path and module resolution issues ## Debugging Tools & Techniques ### Built-in Debugging - Use `print()` statements for quick debugging - Leverage Python's `pdb` debugger for interactive debugging - Add logging with appropriate levels (debug, info, warning, error) - Use `traceback` module for detailed error information ### Testing & Validation - Create minimal reproducible examples - Write unit tests to isolate issues - Use `pytest` with verbose output - Mock external dependencies to isolate problems ### MCP Specific Debugging - Test tools independently before LLM integration - Use `uv run mcp dev server.py` for development testing - Validate schemas with different parameter types - Check Claude Desktop integration logs ## Resolution Approach 1. **Reproduce**: Create a minimal case that demonstrates the issue 2. **Isolate**: Narrow down to the specific component or function 3. **Analyze**: Understand the root cause of the problem 4. **Fix**: Implement a targeted solution 5. **Test**: Verify the fix works and doesn't break other functionality 6. **Document**: Update code comments or documentation if needed Provide the error message, stack trace, or describe the issue you're experiencing, and I'll help debug it systematically.

{ "prompt": "# Debug Python Issues\n\nSystematic debugging assistance for Python code issues, focusing on MCP server development and AI/ML experimentation.\n\n## Debugging Process\n\n### 1. Problem Identification\n- Gather error messages, stack traces, and symptoms\n- Identify when the issue occurs (startup, runtime, specific conditions)\n- Determine affected components (MCP tools, server, dependencies)\n- Check recent changes that might have introduced the issue\n\n### 2. Investigation Strategy\n- **Error Analysis**: Parse stack traces and error messages\n- **Code Review**: Examine relevant code sections for issues\n- **Environment Check**: Verify Python version, dependencies, and configuration\n- **Logging Analysis**: Review application logs and debug output\n\n### 3. Common Issue Categories\n\n#### MCP Server Issues\n- Tool registration and schema generation problems\n- Transport configuration and connection issues\n- Type hint errors affecting schema validation\n- Async/await usage problems\n\n#### Python Code Issues\n- Import errors and dependency conflicts\n- Type annotation problems\n- Exception handling and error propagation\n- Performance bottlenecks and memory issues\n\n#### Environment Issues\n- Conda environment configuration problems\n- Package version conflicts\n- Missing dependencies or incorrect installations\n- Path and module resolution issues\n\n## Debugging Tools & Techniques\n\n### Built-in Debugging\n- Use `print()` statements for quick debugging\n- Leverage Python's `pdb` debugger for interactive debugging\n- Add logging with appropriate levels (debug, info, warning, error)\n- Use `traceback` module for detailed error information\n\n### Testing & Validation\n- Create minimal reproducible examples\n- Write unit tests to isolate issues\n- Use `pytest` with verbose output\n- Mock external dependencies to isolate problems\n\n### MCP Specific Debugging\n- Test tools independently before LLM integration\n- Use `uv run mcp dev server.py` for development testing\n- Validate schemas with different parameter types\n- Check Claude Desktop integration logs\n\n## Resolution Approach\n1. **Reproduce**: Create a minimal case that demonstrates the issue\n2. **Isolate**: Narrow down to the specific component or function\n3. **Analyze**: Understand the root cause of the problem\n4. **Fix**: Implement a targeted solution\n5. **Test**: Verify the fix works and doesn't break other functionality\n6. **Document**: Update code comments or documentation if needed\n\nProvide the error message, stack trace, or describe the issue you're experiencing, and I'll help debug it systematically.", "variables": [], "notes": "Debug Python code issues and errors Tools: codebase, search.", "metadata": { "source": { "provider": "github-prompts", "repository": "https://github.com/yubo-yue/yubo-python", "path": ".github/prompts/debug-issue.prompt.md", "ref": "a6e57136f602a2f4abcbd21eef8844656e6331b0", "url": "https://github.com/yubo-yue/yubo-python/blob/a6e57136f602a2f4abcbd21eef8844656e6331b0/.github/prompts/debug-issue.prompt.md", "key": "yubo-yue/yubo-python/.github/prompts/debug-issue.prompt.md" } }, "model_hints": [ "Claude Sonnet 4" ] }

debug-issue

About