✨ Shopping Agent MCP Tool #95
Conversation
Signed-off-by: Yena Yu <yenayu@bu.edu>
yenayuu
force-pushed
the
add-shopping-agent-mcp
branch
from
e29da6f to
831211f
Compare
|
I see that the PR includes instructions for using the tool. The instructions require an OpenAI key and a SerpAPI key. Can this be run without those keys? |
no it would need those two API Keys to run |
|
@yenayuu, once we are done with this PR, please open a new issue, similar to this one: kagenti/kagenti#405 so we can add this Agent to our Demo collection |
There was a problem hiding this comment.
Pull Request Overview
This PR introduces a new Shopping Agent MCP tool that provides AI-powered product recommendations using LangChain, LangGraph, OpenAI, and SerpAPI. The tool implements a sophisticated multi-step workflow to understand natural language queries, search for products across retailers, and generate personalized recommendations with reasoning.
Key changes include:
- Implementation of a LangGraph-based agent workflow for intelligent shopping recommendations
- Integration with OpenAI for natural language understanding and generation
- SerpAPI integration for real-time product search across retailers
- Comprehensive documentation covering architecture, quickstart, and troubleshooting
Reviewed Changes
Copilot reviewed 17 out of 17 changed files in this pull request and generated 4 comments.
Show a summary per file
| File | Description |
|---|---|
| mcp/shopping_agent/shopping_agent.py | Core MCP server implementation with LangGraph workflow and tool definitions |
| mcp/shopping_agent/pyproject.toml | Project dependencies including FastMCP, LangChain, LangGraph, OpenAI, and SerpAPI |
| mcp/shopping_agent/README.md | Comprehensive documentation with features, setup, usage, and troubleshooting |
| mcp/shopping_agent/QUICKSTART.md | Quick start guide with step-by-step setup and usage examples |
| mcp/shopping_agent/ARCHITECTURE.md | Detailed architecture documentation covering components, data flow, and integration |
| mcp/shopping_agent/TROUBLESHOOTING.md | Troubleshooting guide for common errors and testing methods |
| mcp/shopping_agent/Dockerfile | Container configuration for Docker deployment |
| mcp/shopping_agent/test_simple.py | Direct function testing script bypassing MCP protocol |
| mcp/shopping_agent/test_client.py | HTTP client for testing MCP server endpoints |
| mcp/shopping_agent/test_mcp_client.py | MCP SDK-based test client |
| mcp/shopping_agent/test_server.sh | Shell script for server health checks and MCP protocol testing |
| mcp/shopping_agent/simple_test.py | Standalone test demonstrating LangGraph agent without MCP layer |
| mcp/shopping_agent/setup_env.sh | Helper script for environment setup with API keys |
| mcp/shopping_agent/config.example | Example configuration file for environment variables |
| mcp/shopping_agent/init.py | Package initialization exporting main functions |
| mcp/shopping_agent/.gitignore | Git ignore patterns for secrets, Python artifacts, and IDE files |
| mcp/README.md | Updated MCP tools directory documentation including Shopping Agent |
Comments suppressed due to low confidence (1)
mcp/shopping_agent/QUICKSTART.md:1
- The reference to
movie_toolis unclear since this tool is not mentioned elsewhere in the Shopping Agent documentation. Consider removing this reference or adding context about what the movie_tool is.
# Shopping Agent - Quick Start Guide
💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
| parsed_data = json.loads(response.content) | ||
| product = parsed_data.get("product", "product") | ||
| budget = parsed_data.get("budget", "unknown") | ||
| except: |
There was a problem hiding this comment.
Bare except clause catches all exceptions including SystemExit and KeyboardInterrupt. Use specific exception types like except (json.JSONDecodeError, ValueError): to catch only expected errors.
| except: | |
| except (json.JSONDecodeError, ValueError): |
| "description": content, | ||
| "note": "Please refine your search for more specific results" | ||
| }] | ||
| except Exception as e: |
There was a problem hiding this comment.
Consider catching specific exception types like json.JSONDecodeError and ValueError rather than catching all exceptions. This makes error handling more explicit and prevents catching unexpected errors.
| except Exception as e: | |
| except (json.JSONDecodeError, ValueError) as e: |
| parsed_data = json.loads(response.content) | ||
| product = parsed_data.get("product", "product") | ||
| budget = parsed_data.get("budget", "unknown") | ||
| except: |
There was a problem hiding this comment.
Bare except clause catches all exceptions including SystemExit and KeyboardInterrupt. Use specific exception types like except (json.JSONDecodeError, ValueError): to catch only expected errors.
| except: | |
| except (json.JSONDecodeError, ValueError): |
| "description": content, | ||
| "note": "Please refine your search for more specific results" | ||
| }] | ||
| except Exception as e: |
There was a problem hiding this comment.
Consider catching specific exception types like json.JSONDecodeError and ValueError rather than catching all exceptions. This makes error handling more explicit and prevents catching unexpected errors.
| except Exception as e: | |
| except (json.JSONDecodeError, ValueError) as e: |
3 participants
Summary
Created MCP shopping agent tool
Related issue(s)
Fixes #