22a3fb91ed
## Major Features Added ### 🎯 Dynamic Configuration System - **Environment-aware model selection**: DEFAULT_MODEL with 'pro'/'flash' shortcuts - **Configurable thinking modes**: DEFAULT_THINKING_MODE_THINKDEEP for extended reasoning - **All tool schemas now dynamic**: Show actual current defaults instead of hardcoded values - **Enhanced setup workflow**: Copy from .env.example with smart customization ### 🔧 Model & Thinking Configuration - **Smart model resolution**: Support both shortcuts ('pro', 'flash') and full model names - **Thinking mode optimization**: Only apply thinking budget to models that support it - **Flash model compatibility**: Works without thinking config, still beneficial via system prompts - **Dynamic schema descriptions**: Tool parameters show current environment values ### 🚀 Enhanced Developer Experience - **Fail-fast Docker setup**: GEMINI_API_KEY required upfront in docker-compose - **Comprehensive startup logging**: Shows current model and thinking mode defaults - **Enhanced get_version tool**: Reports all dynamic configuration values - **Better .env documentation**: Clear token consumption details and model options ### 🧪 Comprehensive Testing - **Live model validation**: New simulator test validates Pro vs Flash thinking behavior - **Dynamic configuration tests**: Verify environment variable overrides work correctly - **Complete test coverage**: All 139 unit tests pass, including new model config tests ### 📋 Configuration Files Updated - **docker-compose.yml**: Fail-fast API key validation, thinking mode support - **setup-docker.sh**: Copy from .env.example instead of manual creation - **.env.example**: Detailed documentation with token consumption per thinking mode - **.gitignore**: Added test-setup/ for cleanup ### 🛠 Technical Improvements - **Removed setup.py**: Fully Docker-based deployment (no longer needed) - **REDIS_URL smart defaults**: Auto-configured for Docker, still configurable for dev - **All tools updated**: Consistent dynamic model parameter descriptions - **Enhanced error handling**: Better model resolution and validation ## Breaking Changes - Removed setup.py (Docker-only deployment) - Model parameter descriptions now show actual defaults (dynamic) ## Migration Guide - Update .env files using new .env.example format - Use 'pro'/'flash' shortcuts or full model names - Set DEFAULT_THINKING_MODE_THINKDEEP for custom thinking depth 🤖 Generated with [Claude Code](https://claude.ai/code) Co-Authored-By: Claude <noreply@anthropic.com>
68 lines
3.2 KiB
Python
68 lines
3.2 KiB
Python
"""
|
|
Configuration and constants for Gemini MCP Server
|
|
|
|
This module centralizes all configuration settings for the Gemini MCP Server.
|
|
It defines model configurations, token limits, temperature defaults, and other
|
|
constants used throughout the application.
|
|
|
|
Configuration values can be overridden by environment variables where appropriate.
|
|
"""
|
|
|
|
import os
|
|
|
|
# Version and metadata
|
|
# These values are used in server responses and for tracking releases
|
|
# IMPORTANT: This is the single source of truth for version and author info
|
|
__version__ = "3.3.0" # Semantic versioning: MAJOR.MINOR.PATCH
|
|
__updated__ = "2025-06-11" # Last update date in ISO format
|
|
__author__ = "Fahad Gilani" # Primary maintainer
|
|
|
|
# Model configuration
|
|
# DEFAULT_MODEL: The default model used for all AI operations
|
|
# This should be a stable, high-performance model suitable for code analysis
|
|
# Can be overridden by setting DEFAULT_MODEL environment variable
|
|
DEFAULT_MODEL = os.getenv("DEFAULT_MODEL", "gemini-2.5-pro-preview-06-05")
|
|
|
|
# Token allocation for Gemini Pro (1M total capacity)
|
|
# MAX_CONTEXT_TOKENS: Total model capacity
|
|
# MAX_CONTENT_TOKENS: Available for prompts, conversation history, and files
|
|
# RESPONSE_RESERVE_TOKENS: Reserved for model response generation
|
|
MAX_CONTEXT_TOKENS = 1_000_000 # 1M tokens total capacity for Gemini Pro
|
|
MAX_CONTENT_TOKENS = 800_000 # 800K tokens for content (prompts + files + history)
|
|
RESPONSE_RESERVE_TOKENS = 200_000 # 200K tokens reserved for response generation
|
|
|
|
# Temperature defaults for different tool types
|
|
# Temperature controls the randomness/creativity of model responses
|
|
# Lower values (0.0-0.3) produce more deterministic, focused responses
|
|
# Higher values (0.7-1.0) produce more creative, varied responses
|
|
|
|
# TEMPERATURE_ANALYTICAL: Used for tasks requiring precision and consistency
|
|
# Ideal for code review, debugging, and error analysis where accuracy is critical
|
|
TEMPERATURE_ANALYTICAL = 0.2 # For code review, debugging
|
|
|
|
# TEMPERATURE_BALANCED: Middle ground for general conversations
|
|
# Provides a good balance between consistency and helpful variety
|
|
TEMPERATURE_BALANCED = 0.5 # For general chat
|
|
|
|
# TEMPERATURE_CREATIVE: Higher temperature for exploratory tasks
|
|
# Used when brainstorming, exploring alternatives, or architectural discussions
|
|
TEMPERATURE_CREATIVE = 0.7 # For architecture, deep thinking
|
|
|
|
# Thinking Mode Defaults
|
|
# DEFAULT_THINKING_MODE_THINKDEEP: Default thinking depth for extended reasoning tool
|
|
# Higher modes use more computational budget but provide deeper analysis
|
|
DEFAULT_THINKING_MODE_THINKDEEP = os.getenv("DEFAULT_THINKING_MODE_THINKDEEP", "high")
|
|
|
|
# MCP Protocol Limits
|
|
# MCP_PROMPT_SIZE_LIMIT: Maximum character size for prompts sent directly through MCP
|
|
# The MCP protocol has a combined request+response limit of ~25K tokens.
|
|
# To ensure we have enough space for responses, we limit direct prompt input
|
|
# to 50K characters (roughly ~10-12K tokens). Larger prompts must be sent
|
|
# as files to bypass MCP's token constraints.
|
|
MCP_PROMPT_SIZE_LIMIT = 50_000 # 50K characters
|
|
|
|
# Threading configuration
|
|
# Simple Redis-based conversation threading for stateless MCP environment
|
|
# Set REDIS_URL environment variable to connect to your Redis instance
|
|
REDIS_URL = os.getenv("REDIS_URL", "redis://localhost:6379/0")
|