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

feat: add web search capability to all tools for enhanced analysis

Browse Source 
- Add use_websearch parameter to base ToolRequest class
- All tools now inherit web search capability automatically
- Enhanced description emphasizes brainstorming and architectural design
- Add get_websearch_instruction() helper method to base class
- Update all tools to include web search in their prompts when enabled
- Update README documentation with web search examples
- Fix linting issues (ruff, black formatting)

Web search is particularly useful for:
- Brainstorming sessions and architectural design discussions
- Exploring industry best practices
- Working with specific frameworks/technologies
- Researching solutions to complex problems
- Getting current documentation and community insights

🤖 Generated with Claude Code

Co-Authored-By: Claude <noreply@anthropic.com>
This commit is contained in:
Fahad
2025-06-10 11:09:28 +04:00
parent 2fe83dd381
commit cf0b3f7f44
11 changed files with 191 additions and 15 deletions
Download Patch File Download Diff File Expand all files Collapse all files

46
README.md
View File

@@ -295,11 +295,30 @@ Completely quit and restart Claude Desktop for the changes to take effect.
### 5. Connect to Claude Code
To use the server in Claude Code, run:
#### If you have Claude Desktop installed:
```bash
claude mcp add-from-claude-desktop -s user
```
#### If you only have Claude Code (no desktop app):
**For Traditional Setup (macOS/Linux):**
```bash
claude mcp add gemini -s user -e GEMINI_API_KEY=your-gemini-api-key-here -- /path/to/gemini-mcp-server/run_gemini.sh
```
**For Traditional Setup (Windows):**
```bash
claude mcp add gemini -s user -e GEMINI_API_KEY=your-gemini-api-key-here -- C:\path\to\gemini-mcp-server\run_gemini.bat
```
**For Docker Setup:**
```bash
claude mcp add gemini -s user -- docker run --rm -i --env-file /path/to/gemini-mcp-server/.env -v /home:/workspace:ro gemini-mcp-server:latest
```
Replace `/path/to/gemini-mcp-server` with the actual path where you cloned the repository.
### 6. Start Using It!
Just ask Claude naturally:
@@ -486,6 +505,7 @@ Combine both perspectives to create a comprehensive caching implementation guide
- Architecture and design discussions
- Can reference files for context: `"Use gemini to explain this algorithm with context from algorithm.py"`
- **Dynamic collaboration**: Gemini can request additional files or context during the conversation if needed for a more thorough response
- **Web search capability**: Can search for current documentation, examples, and community solutions when exploring new technologies
**Triggers:** ask, explain, compare, suggest, what about, brainstorm, discuss, share my thinking, get opinion
@@ -503,6 +523,12 @@ Combine both perspectives to create a comprehensive caching implementation guide
"Use gemini to extend my analysis of this distributed system architecture"
```
**With Web Search (for exploring new technologies):**
```
"Use gemini to think deeper about using HTMX vs React for this project - enable web search to explore current best practices"
"Get gemini to think deeper about implementing WebAuthn authentication with web search enabled for latest standards"
```
**Managing Token Costs:**
```
# Claude will intelligently select the right mode, but you can override:
@@ -534,6 +560,7 @@ about event ordering and failure scenarios. Then integrate gemini's insights and
- Validates architectural decisions and design patterns
- Can reference specific files for context: `"Use gemini to think deeper about my API design with reference to api/routes.py"`
- **Enhanced Critical Evaluation (v2.10.0)**: After Gemini's analysis, Claude is prompted to critically evaluate the suggestions, consider context and constraints, identify risks, and synthesize a final recommendation - ensuring a balanced, well-considered solution
- **Web search capability**: When enabled with `use_websearch`, can research current documentation, similar issues, and best practices to inform deeper analysis
**Triggers:** think deeper, ultrathink, extend my analysis, validate my approach
@@ -652,6 +679,12 @@ implementations or missing test coverage. Update the code based on gemini's find
"Get gemini to debug why my API returns 500 errors with the full stack trace: [paste traceback]"
```
**With Web Search (for unfamiliar errors):**
```
"Use gemini to debug this cryptic Kubernetes error with web search enabled to find similar issues"
"Debug this React hydration error with gemini - enable web search to check for known solutions"
```
**Managing Token Costs:**
```
# Save tokens for simple errors
@@ -684,6 +717,7 @@ suggest preventive measures."
- Supports runtime info and previous attempts
- Provides structured root cause analysis with validation steps
- Can request additional context when needed for thorough analysis
- **Web search capability**: When enabled with `use_websearch`, can search for exact error messages, known issues, workarounds, and version-specific problems
**Triggers:** debug, error, failing, root cause, trace, not working
@@ -700,6 +734,12 @@ suggest preventive measures."
"Get gemini to do an architecture analysis of the src/ directory"
```
**With Web Search (for unfamiliar code):**
```
"Use gemini to analyze this GraphQL schema with web search enabled to understand best practices"
"Analyze this Rust code with gemini - enable web search to look up unfamiliar patterns and idioms"
```
**Managing Token Costs:**
```
# Save tokens for quick overviews
@@ -729,6 +769,7 @@ Combine your findings with gemini's to create a comprehensive security report."
- Supports specialized analysis types: architecture, performance, security, quality
- Uses file paths (not content) for clean terminal output
- Can identify patterns, anti-patterns, and refactoring opportunities
- **Web search capability**: When enabled with `use_websearch`, can look up framework documentation, design patterns, and best practices relevant to the code being analyzed
**Triggers:** analyze, examine, look at, understand, inspect
@@ -750,6 +791,7 @@ All tools that work with files support **both individual files and entire direct
- `analysis_type`: architecture|performance|security|quality|general
- `output_format`: summary|detailed|actionable
- `thinking_mode`: minimal|low|medium|high|max (default: medium)
- `use_websearch`: Enable web search for documentation and best practices (default: false)
```
"Use gemini to analyze the src/ directory for architectural patterns"
@@ -776,6 +818,7 @@ All tools that work with files support **both individual files and entire direct
- `runtime_info`: Environment details
- `previous_attempts`: What you've tried
- `thinking_mode`: minimal|low|medium|high|max (default: medium)
- `use_websearch`: Enable web search for error messages and solutions (default: false)
```
"Use gemini to debug this error with context from the entire backend/ directory"
@@ -787,6 +830,7 @@ All tools that work with files support **both individual files and entire direct
- `focus_areas`: Specific aspects to focus on
- `files`: Files or directories for context
- `thinking_mode`: minimal|low|medium|high|max (default: max)
- `use_websearch`: Enable web search for documentation and insights (default: false)
```
"Use gemini to think deeper about my design with reference to the src/models/ directory"

2
config.py
View File

@@ -10,7 +10,7 @@ Configuration values can be overridden by environment variables where appropriat
# Version and metadata
# These values are used in server responses and for tracking releases
__version__ = "2.11.0" # Semantic versioning: MAJOR.MINOR.PATCH
__version__ = "2.11.1" # Semantic versioning: MAJOR.MINOR.PATCH
__updated__ = "2025-06-10" # Last update date in ISO format
__author__ = "Fahad Gilani" # Primary maintainer

2
setup.py
View File

@@ -14,7 +14,7 @@ if readme_path.exists():
setup(
name="gemini-mcp-server",
version="2.8.0",
version="2.11.1",
description="Model Context Protocol server for Google Gemini",
long_description=long_description,
long_description_content_type="text/markdown",

17
tools/analyze.py
View File

@@ -84,6 +84,11 @@ class AnalyzeTool(BaseTool):
"enum": ["minimal", "low", "medium", "high", "max"],
"description": "Thinking depth: minimal (128), low (2048), medium (8192), high (16384), max (32768)",
},
"use_websearch": {
"type": "boolean",
"description": "Enable web search for documentation, best practices, and current information. Particularly useful for: brainstorming sessions, architectural design discussions, exploring industry best practices, working with specific frameworks/technologies, researching solutions to complex problems, or when current documentation and community insights would enhance the analysis.",
"default": False,
},
},
"required": ["files", "question"],
}
@@ -150,10 +155,20 @@ class AnalyzeTool(BaseTool):
focus_instruction = "\n".join(analysis_focus) if analysis_focus else ""
# Add web search instruction if enabled
websearch_instruction = self.get_websearch_instruction(
request.use_websearch,
"""Specifically search for:
- Documentation for technologies or frameworks found in the code
- Best practices and design patterns relevant to the analysis
- API references and usage examples
- Known issues or solutions for patterns you identify""",
)
# Combine everything
full_prompt = f"""{self.get_system_prompt()}
{focus_instruction}
{focus_instruction}{websearch_instruction}
=== USER QUESTION ===
{request.question}

33
tools/base.py
View File

@@ -46,6 +46,10 @@ class ToolRequest(BaseModel):
None,
description="Thinking depth: minimal (128), low (2048), medium (8192), high (16384), max (32768)",
)
use_websearch: Optional[bool] = Field(
False,
description="Enable web search for documentation, best practices, and current information. Particularly useful for: brainstorming sessions, architectural design discussions, exploring industry best practices, working with specific frameworks/technologies, researching solutions to complex problems, or when current documentation and community insights would enhance the analysis.",
)
class BaseTool(ABC):
@@ -145,6 +149,35 @@ class BaseTool(ABC):
"""
return "medium" # Default to medium thinking for better reasoning
def get_websearch_instruction(self, use_websearch: bool, tool_specific: Optional[str] = None) -> str:
"""
Generate web search instruction based on the use_websearch parameter.
Args:
use_websearch: Whether web search is enabled
tool_specific: Optional tool-specific search guidance
Returns:
str: Web search instruction to append to prompt, or empty string
"""
if not use_websearch:
return ""
base_instruction = """
WEB SEARCH ENABLED: Feel free to perform appropriate web searches to enhance your analysis."""
if tool_specific:
return f"{base_instruction} {tool_specific}"
# Default instruction for all tools
return f"""{base_instruction} Search for:
- Current documentation and best practices
- Similar issues and community solutions
- API references and usage examples
- Recent developments and updates
Consider relevant findings when formulating your response."""
@abstractmethod
def get_request_model(self):
"""

17
tools/chat.py
View File

@@ -68,6 +68,11 @@ class ChatTool(BaseTool):
"enum": ["minimal", "low", "medium", "high", "max"],
"description": "Thinking depth: minimal (128), low (2048), medium (8192), high (16384), max (32768)",
},
"use_websearch": {
"type": "boolean",
"description": "Enable web search for documentation, best practices, and current information. Particularly useful for: brainstorming sessions, architectural design discussions, exploring industry best practices, working with specific frameworks/technologies, researching solutions to complex problems, or when current documentation and community insights would enhance the analysis.",
"default": False,
},
},
"required": ["prompt"],
}
@@ -115,8 +120,18 @@ class ChatTool(BaseTool):
# Check token limits
self._validate_token_limit(user_content, "Content")
# Add web search instruction if enabled
websearch_instruction = self.get_websearch_instruction(
request.use_websearch,
"""Specifically search for:
- Documentation for any technologies or concepts mentioned
- Current best practices and patterns
- Recent developments or updates
- Community discussions and solutions""",
)
# Combine system prompt with user content
full_prompt = f"""{self.get_system_prompt()}
full_prompt = f"""{self.get_system_prompt()}{websearch_instruction}
=== USER REQUEST ===
{user_content}

18
tools/debug_issue.py
View File

@@ -82,6 +82,11 @@ class DebugIssueTool(BaseTool):
"enum": ["minimal", "low", "medium", "high", "max"],
"description": "Thinking depth: minimal (128), low (2048), medium (8192), high (16384), max (32768)",
},
"use_websearch": {
"type": "boolean",
"description": "Enable web search for documentation, best practices, and current information. Particularly useful for: brainstorming sessions, architectural design discussions, exploring industry best practices, working with specific frameworks/technologies, researching solutions to complex problems, or when current documentation and community insights would enhance the analysis.",
"default": False,
},
},
"required": ["error_description"],
}
@@ -154,8 +159,19 @@ class DebugIssueTool(BaseTool):
# Check token limits
self._validate_token_limit(full_context, "Context")
# Add web search instruction if enabled
websearch_instruction = self.get_websearch_instruction(
request.use_websearch,
"""Specifically search for:
- The exact error message to find known solutions
- Framework-specific error codes and their meanings
- Similar issues in forums, GitHub issues, or Stack Overflow
- Workarounds and patches for known bugs
- Version-specific issues and compatibility problems""",
)
# Combine everything
full_prompt = f"""{self.get_system_prompt()}
full_prompt = f"""{self.get_system_prompt()}{websearch_instruction}
{full_context}

26
tools/review_changes.py
View File

@@ -92,7 +92,15 @@ class ReviewChanges(BaseTool):
)
def get_input_schema(self) -> dict[str, Any]:
return self.get_request_model().model_json_schema()
schema = self.get_request_model().model_json_schema()
# 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"] = {
"type": "boolean",
"description": "Enable web search for documentation, best practices, and current information. Particularly useful for: brainstorming sessions, architectural design discussions, exploring industry best practices, working with specific frameworks/technologies, researching solutions to complex problems, or when current documentation and community insights would enhance the analysis.",
"default": False,
}
return schema
def get_system_prompt(self) -> str:
return REVIEW_CHANGES_PROMPT
@@ -369,6 +377,17 @@ class ReviewChanges(BaseTool):
)
prompt_parts.extend(context_files_content)
# Add web search instruction if enabled
websearch_instruction = self.get_websearch_instruction(
request.use_websearch,
"""Specifically search for:
- Best practices for new features or patterns introduced
- Security implications of the changes
- Known issues with libraries or APIs being used
- Migration guides if updating dependencies
- Performance considerations for the implemented approach""",
)
# Add review instructions
prompt_parts.append("\n## Review Instructions\n")
prompt_parts.append(
@@ -385,7 +404,10 @@ class ReviewChanges(BaseTool):
"you may request them using the standardized JSON response format."
)
return "\n".join(prompt_parts)
# Combine with system prompt and websearch instruction
full_prompt = f"{self.get_system_prompt()}{websearch_instruction}\n\n" + "\n".join(prompt_parts)
return full_prompt
def format_response(self, response: str, request: ReviewChangesRequest) -> str:
"""Format the response with commit guidance"""

18
tools/review_code.py
View File

@@ -122,6 +122,11 @@ class ReviewCodeTool(BaseTool):
"enum": ["minimal", "low", "medium", "high", "max"],
"description": "Thinking depth: minimal (128), low (2048), medium (8192), high (16384), max (32768)",
},
"use_websearch": {
"type": "boolean",
"description": "Enable web search for documentation, best practices, and current information. Particularly useful for: brainstorming sessions, architectural design discussions, exploring industry best practices, working with specific frameworks/technologies, researching solutions to complex problems, or when current documentation and community insights would enhance the analysis.",
"default": False,
},
},
"required": ["files", "context"],
}
@@ -206,8 +211,19 @@ class ReviewCodeTool(BaseTool):
focus_instruction = "\n".join(review_focus) if review_focus else ""
# Add web search instruction if enabled
websearch_instruction = self.get_websearch_instruction(
request.use_websearch,
"""Specifically search for:
- Security vulnerabilities and CVEs for libraries/frameworks used
- Best practices for the languages and frameworks in the code
- Common anti-patterns and their solutions
- Performance optimization techniques
- Recent updates or deprecations in APIs used""",
)
# Construct the complete prompt with system instructions and code
full_prompt = f"""{self.get_system_prompt()}
full_prompt = f"""{self.get_system_prompt()}{websearch_instruction}
=== USER CONTEXT ===
{request.context}

17
tools/think_deeper.py
View File

@@ -83,6 +83,11 @@ class ThinkDeeperTool(BaseTool):
"description": "Thinking depth: minimal (128), low (2048), medium (8192), high (16384), max (32768)",
"default": "high",
},
"use_websearch": {
"type": "boolean",
"description": "Enable web search for documentation, best practices, and current information. Particularly useful for: brainstorming sessions, architectural design discussions, exploring industry best practices, working with specific frameworks/technologies, researching solutions to complex problems, or when current documentation and community insights would enhance the analysis.",
"default": False,
},
},
"required": ["current_analysis"],
}
@@ -148,8 +153,18 @@ class ThinkDeeperTool(BaseTool):
areas = ", ".join(request.focus_areas)
focus_instruction = f"\n\nFOCUS AREAS: Please pay special attention to {areas} aspects."
# Add web search instruction if enabled
websearch_instruction = self.get_websearch_instruction(
request.use_websearch,
"""Specifically search for:
- Current documentation for technologies, frameworks, or APIs being discussed
- Similar issues and their solutions from the community
- Best practices and recent developments
- Official sources to verify information""",
)
# Combine system prompt with context
full_prompt = f"""{self.get_system_prompt()}{focus_instruction}
full_prompt = f"""{self.get_system_prompt()}{focus_instruction}{websearch_instruction}
{full_context}

10
utils/git_utils.py
View File

@@ -52,7 +52,7 @@ def find_git_repositories(start_path: str, max_depth: int = 5) -> list[str]:
List of absolute paths to git repositories, sorted alphabetically
"""
repositories = []
try:
# Create Path object - no need to resolve yet since the path might be
# a translated Docker path that doesn't exist on the host
@@ -61,12 +61,12 @@ def find_git_repositories(start_path: str, max_depth: int = 5) -> list[str]:
# Basic validation - must be absolute
if not start_path.is_absolute():
return []
# Check if the path exists before trying to walk it
if not start_path.exists():
return []
except Exception as e:
except Exception:
# If there's any issue with the path, return empty list
return []
@@ -122,7 +122,7 @@ def run_git_command(repo_path: str, command: list[str]) -> tuple[bool, str]:
# Verify the repository path exists before trying to use it
if not Path(repo_path).exists():
return False, f"Repository path does not exist: {repo_path}"
try:
# Execute git command with safety measures
result = subprocess.run(