How to debug: Invalid arguments for firecrawl_extract Error

[eboraks]

I am being played around with the Firecrawl MCP server, and I need help understanding how to debug the following error:

[error] {"message":"Request failed: Invalid arguments for firecrawl_extract","tool":"firecrawl_extract","arguments":{"url":"https://www.womenshistory.org/education-resources/biographies/margaret-sanger","extraction_schema":{"type":"object","properties":{"article_title":{"type":"string","description":"The title of the article"},"article_text":{"type":"string","description":"The full text of the article"}},"required":["article_title","article_text"]}},"timestamp":"2025-05-18T13:44:05.403Z","duration":16} [info] Request completed in 17ms 2025-05-18 09:44:05,408 - mcp_use - ERROR - Error parsing tool result: Tool execution failed: [TextContent(type='text', text='Error: Invalid arguments for firecrawl_extract', annotations=None)]

My set-up & code:

import os
from langchain_google_genai import ChatGoogleGenerativeAI
from mcp_use import MCPAgent, MCPClient

llm = ChatGoogleGenerativeAI(api_key=os.getenv("GOOGLE_API_KEY"), model=os.getenv("GEMINI_PRO_MODEL"))

firecrawl_api_key = os.getenv("FIRECRAWL_API_KEY")
notion_api_key = os.getenv("NOTION_API_KEY")


config = {
    "mcpServers": {
        "firecrawl": {
        "command": "docker",
        "args": [
            "run",
            "-i",
            "--rm",
            "--name",
            "firecrawl-container",
            "-e",
            "FIRECRAWL_API_URL",
            "-e",
            "FIRECRAWL_RETRY_MAX_ATTEMPTS",
            "-e",
            "FIRECRAWL_RETRY_INITIAL_DELAY",
            "-e",
            "FIRECRAWL_RETRY_MAX_DELAY",
            "-e",
            "FIRECRAWL_RETRY_BACKOFF_FACTOR",
            "-e",
            "FIRECRAWL_CREDIT_WARNING_THRESHOLD",
            "-e",
            "FIRECRAWL_CREDIT_CRITICAL_THRESHOLD",
            "-e",
            "FIRECRAWL_API_KEY",
            "mcp/firecrawl"
        ],
        "env": {
            "FIRECRAWL_API_URL": "https://api.firecrawl.dev/v1",
            "FIRECRAWL_RETRY_MAX_ATTEMPTS": "5",
            "FIRECRAWL_RETRY_INITIAL_DELAY": "2000",
            "FIRECRAWL_RETRY_MAX_DELAY": "30000",
            "FIRECRAWL_RETRY_BACKOFF_FACTOR": "3",
            "FIRECRAWL_CREDIT_WARNING_THRESHOLD": "2000",
            "FIRECRAWL_CREDIT_CRITICAL_THRESHOLD": "500",
            "FIRECRAWL_API_KEY": "YOUR-API-KEY"
        }
        },
        "Wikipedia": {
            "command": "npx",
            "args": ["-y", "wikipedia-mcp"]
        },
        "notionApi": {
            "command": "npx",
            "args": ["-y", "@notionhq/notion-mcp-server"],
            "env": {
                "OPENAPI_MCP_HEADERS": "{\"Authorization\": \"Bearer " + notion_api_key + "\", \"Notion-Version\": \"2022-06-28\" }"
            }
        }
    }
}

# Create MCPClient from configuration dictionary
client = MCPClient.from_dict(config)

# Create agent with the client
agent = MCPAgent(llm=llm, client=client, max_steps=30, use_server_manager=True, verbose=True)

# Run the query
result = asyncio.run(agent.run(
    """Search wikipedia for 'Margaret Sanger' and tell me what I need to know about her. 
    the user firecrawl to open https://www.womenshistory.org/education-resources/biographies/margaret-sanger 
    and extract the text from the article and print the title of the article."""
))
print(f"\nResult: {result}")```