CSV Agent does not pass parameter to allow dangerous code (1.0.6)

[Arron-Clague]

CSV Agent does not pass parameter to allow dangerous code to langchain component, and throws an error.

Suggested change to CSV Component :

from langchain_experimental.agents.agent_toolkits.csv.base import create_csv_agent

from langflow.custom import CustomComponent from langflow.field_typing import AgentExecutor, LanguageModel

class CSVAgentComponent(CustomComponent): display_name = "CSVAgent" description = "Construct a CSV agent from a CSV and tools - ADC." documentation = "https://python.langchain.com/docs/modules/agents/toolkits/csv"

def build_config(self):
    return {
        "llm": {"display_name": "LLM", "type": LanguageModel},
        "path": {"display_name": "Path", "field_type": "file", "suffixes": [".csv"], "file_types": [".csv"]},
        "adc": {"display_name": "Allow Dangerous Code", "type": True,"advanced" : True},
        "handle_parsing_errors": {"display_name": "Handle Parse Errors", "advanced": True},
        "agent_type": {
            "display_name": "Agent Type",
            "options": ["zero-shot-react-description", "openai-functions", "openai-tools"],
            "advanced": True,
        },
    }

def build(
    self, llm: LanguageModel, path: str, handle_parsing_errors: bool = True, agent_type: str = "openai-tools", adc: bool=True
) -> AgentExecutor:
    # Instantiate and return the CSV agent class with the provided llm and path
    return create_csv_agent(
        llm=llm,
        path=path,
        agent_type=agent_type,
        verbose=True,
        allow_dangerous_code=adc,
        agent_executor_kwargs=dict(handle_parsing_errors=handle_parsing_errors),
    )

[dosubot[bot]]

To modify the CSV Agent in version 1.0.6 to pass the parameter to allow dangerous code to the langchain component, you can update the build method in the CSVAgentComponent class to include the allow_dangerous_code parameter. Here is the suggested code change:

from langchain_experimental.agents.agent_toolkits.csv.base import create_csv_agent
from langflow import CustomComponent
from langflow.field_typing import BaseLanguageModel, AgentExecutor

class CSVAgentComponent(CustomComponent):
    display_name = "CSVAgent"
    description = "Construct a CSV agent from a CSV and tools."
    documentation = "https://python.langchain.com/docs/modules/agents/toolkits/csv"

    def build_config(self):
        return {
            "llm": {"display_name": "LLM", "type": BaseLanguageModel},
            "path": {"display_name": "Path", "field_type": "file", "suffixes": [".csv"], "file_types": [".csv"]},
            "handle_parsing_errors": {"display_name": "Handle Parse Errors", "advanced": True},
            "agent_type": {
                "display_name": "Agent Type",
                "options": ["zero-shot-react-description", "openai-functions", "openai-tools"],
                "advanced": True,
            },
            "allow_dangerous_code": {"display_name": "Allow Dangerous Code", "advanced": True, "type": "boolean"},
        }

    def build(
        self, llm: BaseLanguageModel, path: str, handle_parsing_errors: bool = True, agent_type: str = "openai-tools", allow_dangerous_code: bool = False
    ) -> AgentExecutor:
        # Instantiate and return the CSV agent class with the provided llm and path
        return create_csv_agent(
            llm=llm,
            path=path,
            agent_type=agent_type,
            verbose=True,
            agent_executor_kwargs=dict(handle_parsing_errors=handle_parsing_errors, allow_dangerous_code=allow_dangerous_code),
        )

This change adds the allow_dangerous_code parameter to the build_config method and includes it in the build method, passing it to the create_csv_agent function ^[1][2][3].

To continue talking to Dosu, mention @dosu.