SOS!When I’m using the knowledge base, I don’t want to do a function calling, I change how to close it

General
, , ,
1

When I’m using the knowledge base, I don’t want to do a function calling, I change how to close it

log content:
INFO Creating collection
INFO Loading knowledge base
ERROR Function open_application not found
ERROR Function open_application not found
ERROR Function open_application not found
ERROR Function launch_application not found
ERROR Function open_application not found
ERROR Function open_application not found
ERROR Function open_application not found

My agent configuration:
agent = Agent(
model=OpenAIChat(base_url=“xxxxxx”, provider=“ollama_chat”, api_key=“ollama”,
id=“xxxxxx”, temperature=0.1),
knowledge=knowledge_base,
search_knowledge=True,
tools=,show_tool_calls=True,
description=“Which plugin should I use with arguments doesn’t require you to execute the functions in the knowledge base”,
)
agent.knowledge.load(recreate=False)

agent.print_response(UserContent)
return agent.run_response.content

1 Like
2

Hey @hope_321
thanks for reaching out and supporting Agno. I’ve shared this with the team, we’re working through all requests one by one and will get back to you soon.
If it’s urgent, please let us know. We appreciate your patience!

3

Hi! We were not able to reproduce this issue.

Here’s a quick example similar to your use case:

from agno.agent import Agent
from agno.models.openai import OpenAIChat
from agno.knowledge.pdf_url import PDFUrlKnowledgeBase
from agno.vectordb.lancedb import LanceDb, SearchType
from agno.embedder.openai import OpenAIEmbedder
from pathlib import Path
from agno.models.ollama import Ollama

# Setup paths
# Use a local path instead of the mapped network drive for LanceDB
tmp_dir = Path("C:\\temp\\agno_lancedb")
tmp_dir.mkdir(parents=True, exist_ok=True)

# Configure the language model
model = Ollama(id="llama3.1:8b")

knowledge_base=PDFUrlKnowledgeBase(
        urls=["https://agno-public.s3.amazonaws.com/recipes/ThaiRecipes.pdf"],
        vector_db=LanceDb(
            uri=str(tmp_dir), # Use the local path directly
            table_name="recipes",
            search_type=SearchType.hybrid,
            embedder=OpenAIEmbedder(id="text-embedding-3-small"),
        ),
    )

# Placeholder for user query
UserContent = "Which plugin helps me look at files on my computer?"

agent = Agent(
    model=model,
    knowledge=knowledge_base,
    search_knowledge=True,
    tools=[],
    show_tool_calls=True,
    description="Which plugin should I use with arguments doesn't require you to execute the functions in the knowledge base",
)
knowledge_base.load(upsert=True)

agent.print_response(UserContent)

Additionally if you still get errors/ tool calls issues reply again and share the entire code block and debug logs.