Skip to main content

Documentation Index

Fetch the complete documentation index at: https://mcp-server-langgraph.mintlify.app/llms.txt

Use this file to discover all available pages before exploring further.

​
AI Tools Compatibility Guide

This codebase is optimized for compatibility with multiple AI coding assistants and IDEs.

​
✅ Fully Supported AI Tools

​
1. Claude Code (Anthropic)

  • Status: ✅ Fully Supported
  • Configuration: Uses .cursorrules and .ai/ directory
  • Features:
    • Project-aware code generation
    • Context-aware refactoring
    • Documentation generation
    • Test generation
  • Best Practices:
    • Reads entire codebase context
    • Follows established patterns
    • Maintains consistency

​
2. GitHub Copilot (Microsoft/OpenAI)

  • Status: ✅ Fully Supported
  • Configuration:
    • .github/copilot-instructions.md
    • .vscode/settings.json
  • Features:
    • Inline code completions
    • Function generation
    • Test suggestions
    • Documentation
  • Best Practices:
    • Use inline comments to guide suggestions
    • Leverage #file: syntax for context
    • Tab through multiple suggestions

​
3. Cursor (Anysphere)

  • Status: ✅ Fully Supported
  • Configuration:
    • .cursorrules
    • .cursor/mcp.json
    • .vscode/* (inherited)
  • Features:
    • Chat with codebase
    • Multi-file edits
    • MCP server integration
    • @-mentions for docs
  • Best Practices:
    • Use @codebase for project-wide questions
    • Use @docs for documentation queries
    • Use @git for history context

​
4. Gemini Code Assist (Google)

  • Status: ✅ Supported
  • Configuration: Uses workspace settings
  • Features:
    • Code completions
    • Refactoring suggestions
    • Documentation generation
  • Best Practices:
    • Integrates with Google Cloud
    • Follows EditorConfig
    • Respects .gitignore

​
5. OpenAI Codex (OpenAI)

  • Status: ✅ Fully Supported
  • Configuration:
    • .openai/codex-instructions.md
    • .openai/code-completion-config.json
    • .vscode/settings.json (OpenAI section)
  • Features:
    • Advanced code completions
    • Multi-line suggestions
    • Context-aware generation
    • Code explanation and documentation
  • Best Practices:
    • Use clear function signatures as prompts
    • Provide context via comments
    • Review generated code for security
    • Test completions thoroughly

​
6. OpenAI GPT-4 (via extensions)

  • Status: ✅ Supported
  • Configuration: Extension-specific, uses .openai/ configs
  • Features:
    • Code generation
    • Explanation
    • Debugging assistance
    • Refactoring suggestions
  • Best Practices:
    • Use prompt templates from .ai/prompts.md
    • Provide context from documentation
    • Leverage .openai/codex-instructions.md

​
🔧 IDE Compatibility

​
Visual Studio Code

  • Status: ✅ Fully Configured
  • Configuration Files:
    • .vscode/settings.json - Editor, Python, formatting
    • .vscode/extensions.json - Recommended extensions
    • .vscode/launch.json - Debugging configurations
    • .vscode/tasks.json - Build and test tasks
  • AI Extensions Supported:
    • GitHub Copilot
    • GitHub Copilot Chat
    • OpenAI Codex
    • Cursor (VSCode fork)
    • Tabnine
    • AWS CodeWhisperer
  • Recommended Extensions: 
    ms-python.python
ms-python.black-formatter
ms-python.isort
github.copilot
github.copilot-chat
ms-azuretools.vscode-docker
ms-kubernetes-tools.vscode-kubernetes-tools

​
Cursor

  • Status: ✅ Fully Configured
  • Configuration Files:
    • .cursorrules - AI behavior rules
    • .cursor/mcp.json - MCP server integration
    • Inherits all .vscode/* configs
  • Unique Features:
    • Native MCP protocol support
    • Multi-file editing
    • Codebase-aware chat
  • Setup:
    1. Open project in Cursor
    2. MCP server auto-detected via .cursor/mcp.json
    3. Use Cmd+K for inline edits
    4. Use Cmd+L for chat

​
JetBrains IDEs (PyCharm, IntelliJ)

  • Status: ⚠️ Partial Support
  • Configuration: .editorconfig provides base settings
  • AI Support:
    • GitHub Copilot plugin
    • AI Assistant (JetBrains)
  • Note: May need manual Python interpreter setup

​
Neovim / Vim

  • Status: ⚠️ Manual Configuration Needed
  • Recommendations:
    • Use coc.nvim with Python LSP
    • Install GitHub Copilot plugin
    • Respect .editorconfig
  • Configuration: User-managed

​
📋 Configuration Files Overview

​
Universal (All Tools)

FilePurposeSupported By
.editorconfigEditor settingsAll IDEs
.gitignoreVCS exclusionsAll tools
pyproject.tomlPython toolingAll Python tools
.pre-commit-config.yamlGit hooksAll VCS workflows

​
AI-Specific

FilePurposeUsed By
.cursorrulesAI behavior rulesCursor, Claude Code
.github/copilot-instructions.mdCopilot guidelinesGitHub Copilot
.openai/codex-instructions.mdOpenAI Codex guideOpenAI Codex, GPT-4
.openai/code-completion-config.jsonCodex configurationOpenAI Codex
.ai/README.mdAI assistant guideAll AI tools
.ai/prompts.mdPrompt templatesAll AI tools

​
IDE-Specific

FilePurposeUsed By
.vscode/settings.jsonEditor configVSCode, Cursor
.vscode/extensions.jsonExtension recommendationsVSCode, Cursor
.vscode/launch.jsonDebug configsVSCode, Cursor
.vscode/tasks.jsonBuild tasksVSCode, Cursor
.cursor/mcp.jsonMCP serverCursor

​
🚀 Quick Start by Tool

​
VSCode with GitHub Copilot

  1. Install VSCode: https://code.visualstudio.com/
  2. Install Extensions: 
    code --install-extension ms-python.python
code --install-extension github.copilot
code --install-extension github.copilot-chat
  3. Open Project: 
    code /path/to/mcp_server_langgraph
  4. Install Recommended Extensions: Click “Install” when prompted
  5. Setup Python:
    • Cmd+Shift+P → “Python: Select Interpreter”
    • Choose ./venv/bin/python

​
Cursor

  1. Install Cursor: https://cursor.sh/
  2. Open Project: 
    cursor /path/to/mcp_server_langgraph
  3. MCP Auto-detected: Cursor reads .cursor/mcp.json automatically
  4. Start Coding:
    • Cmd+K for inline edits
    • Cmd+L for chat
    • @codebase for project context

​
Claude Code (CLI)

  1. Install Claude Code: https://docs.anthropic.com/claude-code
  2. Navigate to Project: 
    cd /path/to/mcp_server_langgraph
  3. Start Session: 
    claude-code
  4. Reads Automatically:
    • .cursorrules
    • .ai/README.md
    • Project structure

​
OpenAI Codex

  1. Install OpenAI Extension:
    • VSCode: Install OpenAI extension
    • Or use OpenAI API directly
  2. Configure API Key: 
    # Add to environment
export OPENAI_API_KEY="sk-..."
  3. Open Project: 
    code /path/to/mcp_server_langgraph
  4. Configuration Auto-loaded:
    • .openai/codex-instructions.md
    • .openai/code-completion-config.json
    • .vscode/settings.json (OpenAI section)

​
Gemini Code Assist

  1. Install Cloud Code: https://cloud.google.com/code
  2. Open Project in Supported IDE:
    • VSCode
    • JetBrains
  3. Enable Gemini:
    • Settings → Cloud Code → AI Assistance
  4. Start Coding: Inline suggestions appear automatically

​
🎯 Feature Comparison

FeatureVSCode+CopilotCursorClaude CodeOpenAI CodexGemini
Inline Completions✅✅✅✅✅
Chat Interface✅✅✅✅✅
Multi-file Edits⚠️✅✅✅⚠️
Codebase Context⚠️✅✅✅⚠️
MCP Integration❌✅✅⚠️❌
Test Generation✅✅✅✅✅
Refactoring✅✅✅✅✅
Documentation✅✅✅✅✅
Function-level Gen✅✅✅✅✅
Security Analysis⚠️✅✅✅⚠️
Legend: ✅ Excellent | ⚠️ Partial | ❌ Not Available

​
🎨 Code Style Consistency

All AI tools respect these configurations:

​
Formatting

  • Black: 127 character line length
  • isort: Black-compatible import sorting
  • flake8: Linting with E203, W503 ignored
  • mypy: Strict type checking

​
Editor Settings

  • Indent: 4 spaces for Python
  • Indent: 2 spaces for YAML, JSON
  • Trailing whitespace: Auto-removed
  • Final newline: Auto-inserted
  • Line endings: LF (Unix)

​
File Exclusions

All tools ignore:
  • venv/, .venv/
  • __pycache__/, *.pyc
  • .pytest_cache/, .mypy_cache/
  • logs/
  • htmlcov/, .coverage

​
📚 Documentation for AI Tools

​
Where AI Tools Should Look First

  1. Project Overview: README.md
  2. Architecture: README.md → Architecture section
  3. Development Setup: DEVELOPMENT.md
  4. Contributing: CONTRIBUTING.md
  5. Code Patterns:
    • .cursorrules (Cursor, Claude)
    • .openai/codex-instructions.md (OpenAI Codex)
    • .github/copilot-instructions.md (GitHub Copilot)
  6. API Reference: http://localhost:8000/docs (when running)

​
Prompt Templates

See .ai/prompts.md for pre-written prompts for common tasks:
  • Adding new tools
  • Creating API endpoints
  • Writing tests
  • Debugging issues
  • Performance optimization
  • Security reviews

​
🔒 Security Considerations

All AI tools are configured to:

​
Never Commit

  • Secrets or API keys
  • .env files (only .env.example)
  • Credentials
  • Private keys

​
Always Validate

  • User inputs with Pydantic
  • JWT tokens
  • Authorization with OpenFGA
  • Environment variables

​
Logging Rules

  • ✅ Log events, errors, metrics
  • ✅ Log user IDs, resource IDs
  • ❌ Never log passwords, tokens, secrets
  • ❌ Never log PII without encryption

​
🧪 Testing with AI Tools

​
Test Generation Prompts

For any AI tool: 
Generate comprehensive tests for [function/class] including:
- Happy path
- Error cases
- Edge cases
- Performance benchmarks (if applicable)

Use pytest markers:
- @pytest.mark.unit
- @pytest.mark.integration
- @pytest.mark.benchmark

Mock external dependencies.

​
Running Tests

All AI tools can suggest: 
# All tests
pytest -v

# Specific marker
pytest -m unit -v

# With coverage
pytest --cov=. --cov-report=html

# Benchmarks
pytest -m benchmark --benchmark-only

​
🐛 Debugging with AI Tools

​
Debug Prompts

I'm seeing this error:
[paste error]

Help me debug by:
1. Explaining the error
2. Identifying root cause
3. Suggesting fixes
4. Adding tests to prevent regression

​
Debug Configurations

VSCode/Cursor users can use preconfigured launch configs:
  • Python: MCP Server (StreamableHTTP)
  • Python: Debug Tests
  • Python: Current File
See .vscode/launch.json for all options.

​
📊 Telemetry and Observability

AI tools understand our observability patterns:

​
Tracing

with tracer.start_as_current_span("operation") as span:
    span.set_attribute("key", "value")
    # operation

​
Logging

logger.info("Event", extra={"context": "data"})

​
Metrics

metrics.counter.add(1, {"label": "value"})
AI tools will suggest adding telemetry to new code.

​
🚢 Deployment Awareness

AI tools understand deployment contexts:
  • Local Development: Docker Compose
  • Staging: Kubernetes with Kustomize
  • Production: Kubernetes with Helm
See KUBERNETES_DEPLOYMENT.md and PRODUCTION_DEPLOYMENT.md.

​
🤝 Contributing AI Configurations

If you improve AI tool configurations:
  1. Test with multiple tools (VSCode, Cursor, Claude)
  2. Document in this file
  3. Update .ai/README.md
  4. Submit PR with examples

​
📞 Support

​
Issues with AI Tools

​
Project-Specific Issues

​
🔄 Keeping Configurations Updated

When updating AI configurations:
  1. Test Changes: Verify with at least 2 AI tools
  2. Update Docs: This file and .ai/README.md
  3. Version Control: Commit with clear message
  4. Notify Team: PR description should explain changes

​
📈 Future Compatibility

We’re monitoring and will add support for:
  • ⏳ Amazon CodeWhisperer - AWS-powered completions
  • ⏳ Tabnine - Privacy-focused AI assistant
  • ⏳ Codeium - Free AI code acceleration
  • ⏳ Sourcegraph Cody - Code intelligence platform
  • ⏳ Replit Ghostwriter - Collaborative AI coding
PRs welcome for additional tool support!

​
Recently Added

  • ✅ OpenAI Codex (2025-10-10) - Full configuration added

​
✅ Verification Checklist

To verify AI tool compatibility:
  • .editorconfig respected
  • .vscode/settings.json loaded
  • Python formatter (Black) working
  • Import sorter (isort) working
  • Linter (flake8) showing errors
  • Type checker (mypy) working
  • Tests runnable
  • Debug configs working
  • AI completions appearing
  • Project context understood
Run this verification in your IDE of choice!
Last Updated: 2025-10-10 Maintained By: @vishnu2kmohan License: MIT