torbjorn/my-pal-mcp-server
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

feat: Add comprehensive dynamic configuration system v3.3.0

Browse Source 
## 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>
This commit is contained in:
Fahad
2025-06-11 20:10:25 +04:00
parent f797480006
commit 22a3fb91ed
18 changed files with 351 additions and 116 deletions
Download Patch File Download Diff File Expand all files Collapse all files

6
tools/analyze.py
View File

@@ -42,6 +42,8 @@ class AnalyzeTool(BaseTool):
)
def get_input_schema(self) -> dict[str, Any]:
from config import DEFAULT_MODEL
return {
"type": "object",
"properties": {
@@ -50,6 +52,10 @@ class AnalyzeTool(BaseTool):
"items": {"type": "string"},
"description": "Files or directories to analyze (must be absolute paths)",
},
"model": {
"type": "string",
"description": f"Model to use: 'pro' (Gemini 2.5 Pro with extended thinking) or 'flash' (Gemini 2.0 Flash - faster). Defaults to '{DEFAULT_MODEL}' if not specified.",
},
"question": {
"type": "string",
"description": "What to analyze or look for",

26
tools/base.py
View File

@@ -25,7 +25,7 @@ from google.genai import types
from mcp.types import TextContent
from pydantic import BaseModel, Field
from config import GEMINI_MODEL, MAX_CONTEXT_TOKENS, MCP_PROMPT_SIZE_LIMIT
from config import DEFAULT_MODEL, MAX_CONTEXT_TOKENS, MCP_PROMPT_SIZE_LIMIT
from utils import check_token_limit
from utils.conversation_memory import (
MAX_CONVERSATION_TURNS,
@@ -50,7 +50,10 @@ class ToolRequest(BaseModel):
these common fields.
"""
model: Optional[str] = Field(None, description="Model to use (defaults to Gemini 2.5 Pro)")
model: Optional[str] = Field(
None,
description=f"Model to use: 'pro' (Gemini 2.5 Pro with extended thinking) or 'flash' (Gemini 2.0 Flash - faster). Defaults to '{DEFAULT_MODEL}' if not specified.",
)
temperature: Optional[float] = Field(None, description="Temperature for response (tool-specific defaults)")
# Thinking mode controls how much computational budget the model uses for reasoning
# Higher values allow for more complex reasoning but increase latency and cost
@@ -625,7 +628,7 @@ If any of these would strengthen your analysis, specify what Claude should searc
# No need to rebuild it here - prompt already contains conversation history
# Extract model configuration from request or use defaults
model_name = getattr(request, "model", None) or GEMINI_MODEL
model_name = getattr(request, "model", None) or DEFAULT_MODEL
temperature = getattr(request, "temperature", None)
if temperature is None:
temperature = self.get_default_temperature()
@@ -1064,13 +1067,22 @@ If any of these would strengthen your analysis, specify what Claude should searc
temperature and thinking budget configuration for models that support it.
Args:
model_name: Name of the Gemini model to use
model_name: Name of the Gemini model to use (or shorthand like 'flash', 'pro')
temperature: Temperature setting for response generation
thinking_mode: Thinking depth mode (affects computational budget)
Returns:
Model instance configured and ready for generation
"""
# Define model shorthands for user convenience
model_shorthands = {
"pro": "gemini-2.5-pro-preview-06-05",
"flash": "gemini-2.0-flash-exp",
}
# Resolve shorthand to full model name
resolved_model_name = model_shorthands.get(model_name.lower(), model_name)
# Map thinking modes to computational budget values
# Higher budgets allow for more complex reasoning but increase latency
thinking_budgets = {
@@ -1085,7 +1097,7 @@ If any of these would strengthen your analysis, specify what Claude should searc
# Gemini 2.5 models support thinking configuration for enhanced reasoning
# Skip special handling in test environment to allow mocking
if "2.5" in model_name and not os.environ.get("PYTEST_CURRENT_TEST"):
if "2.5" in resolved_model_name and not os.environ.get("PYTEST_CURRENT_TEST"):
try:
# Retrieve API key for Gemini client creation
api_key = os.environ.get("GEMINI_API_KEY")
@@ -1144,7 +1156,7 @@ If any of these would strengthen your analysis, specify what Claude should searc
return ResponseWrapper(response.text)
return ModelWrapper(client, model_name, temperature, thinking_budget)
return ModelWrapper(client, resolved_model_name, temperature, thinking_budget)
except Exception:
# Fall back to regular API if thinking configuration fails
@@ -1197,4 +1209,4 @@ If any of these would strengthen your analysis, specify what Claude should searc
return ResponseWrapper(response.text)
return SimpleModelWrapper(client, model_name, temperature)
return SimpleModelWrapper(client, resolved_model_name, temperature)

6
tools/chat.py
View File

@@ -44,6 +44,8 @@ class ChatTool(BaseTool):
)
def get_input_schema(self) -> dict[str, Any]:
from config import DEFAULT_MODEL
return {
"type": "object",
"properties": {
@@ -56,6 +58,10 @@ class ChatTool(BaseTool):
"items": {"type": "string"},
"description": "Optional files for context (must be absolute paths)",
},
"model": {
"type": "string",
"description": f"Model to use: 'pro' (Gemini 2.5 Pro with extended thinking) or 'flash' (Gemini 2.0 Flash - faster). Defaults to '{DEFAULT_MODEL}' if not specified.",
},
"temperature": {
"type": "number",
"description": "Response creativity (0-1, default 0.5)",

6
tools/codereview.py
View File

@@ -79,6 +79,8 @@ class CodeReviewTool(BaseTool):
)
def get_input_schema(self) -> dict[str, Any]:
from config import DEFAULT_MODEL
return {
"type": "object",
"properties": {
@@ -87,6 +89,10 @@ class CodeReviewTool(BaseTool):
"items": {"type": "string"},
"description": "Code files or directories to review (must be absolute paths)",
},
"model": {
"type": "string",
"description": f"Model to use: 'pro' (Gemini 2.5 Pro with extended thinking) or 'flash' (Gemini 2.0 Flash - faster). Defaults to '{DEFAULT_MODEL}' if not specified.",
},
"context": {
"type": "string",
"description": "User's summary of what the code does, expected behavior, constraints, and review objectives",

6
tools/debug.py
View File

@@ -50,6 +50,8 @@ class DebugIssueTool(BaseTool):
)
def get_input_schema(self) -> dict[str, Any]:
from config import DEFAULT_MODEL
return {
"type": "object",
"properties": {
@@ -57,6 +59,10 @@ class DebugIssueTool(BaseTool):
"type": "string",
"description": "Error message, symptoms, or issue description",
},
"model": {
"type": "string",
"description": f"Model to use: 'pro' (Gemini 2.5 Pro with extended thinking) or 'flash' (Gemini 2.0 Flash - faster). Defaults to '{DEFAULT_MODEL}' if not specified.",
},
"error_context": {
"type": "string",
"description": "Stack trace, logs, or additional error context",

8
tools/precommit.py
View File

@@ -98,7 +98,15 @@ class Precommit(BaseTool):
)
def get_input_schema(self) -> dict[str, Any]:
from config import DEFAULT_MODEL
schema = self.get_request_model().model_json_schema()
# Ensure model parameter has enhanced description
if "properties" in schema and "model" in schema["properties"]:
schema["properties"]["model"] = {
"type": "string",
"description": f"Model to use: 'pro' (Gemini 2.5 Pro with extended thinking) or 'flash' (Gemini 2.0 Flash - faster). Defaults to '{DEFAULT_MODEL}' if not specified.",
}
# Ensure use_websearch is in the schema with proper description
if "properties" in schema and "use_websearch" not in schema["properties"]:
schema["properties"]["use_websearch"] = {

15
tools/thinkdeep.py
View File

@@ -48,6 +48,8 @@ class ThinkDeepTool(BaseTool):
)
def get_input_schema(self) -> dict[str, Any]:
from config import DEFAULT_MODEL
return {
"type": "object",
"properties": {
@@ -55,6 +57,10 @@ class ThinkDeepTool(BaseTool):
"type": "string",
"description": "Your current thinking/analysis to extend and validate",
},
"model": {
"type": "string",
"description": f"Model to use: 'pro' (Gemini 2.5 Pro with extended thinking) or 'flash' (Gemini 2.0 Flash - faster). Defaults to '{DEFAULT_MODEL}' if not specified.",
},
"problem_context": {
"type": "string",
"description": "Additional context about the problem or goal",
@@ -78,8 +84,7 @@ class ThinkDeepTool(BaseTool):
"thinking_mode": {
"type": "string",
"enum": ["minimal", "low", "medium", "high", "max"],
"description": "Thinking depth: minimal (128), low (2048), medium (8192), high (16384), max (32768)",
"default": "high",
"description": f"Thinking depth: minimal (128), low (2048), medium (8192), high (16384), max (32768). Defaults to '{self.get_default_thinking_mode()}' if not specified.",
},
"use_websearch": {
"type": "boolean",
@@ -101,8 +106,10 @@ class ThinkDeepTool(BaseTool):
return TEMPERATURE_CREATIVE
def get_default_thinking_mode(self) -> str:
"""ThinkDeep uses high thinking by default"""
return "high"
"""ThinkDeep uses configurable thinking mode, defaults to high"""
from config import DEFAULT_THINKING_MODE_THINKDEEP
return DEFAULT_THINKING_MODE_THINKDEEP
def get_request_model(self):
return ThinkDeepRequest