----------------------
Based on the provided knowledge sources, there is no direct method to add more tools to an instance of
FunctionCallingAgentWorker after it has been created. The
FunctionCallingAgentWorker is initialized with a set of tools or a
tool_retriever, and these are not designed to be modified after the instance has been created.
However, there is a way to work around this by using a
tool_retriever. The
tool_retriever is an instance of
ObjectRetriever that can retrieve tools based on some criteria. If you initialize the
FunctionCallingAgentWorker with a
tool_retriever, you can potentially modify the underlying data of the
tool_retriever to change the tools that are available.
Here is an example of creating a
FunctionCallingAgentWorker with a
tool_retriever:
from llama_index.core import VectorStoreIndex
from llama_index.core.objects import ObjectIndex
# define an "object" index and retriever over these tools
obj_index = ObjectIndex.from_objects(
all_tools,
index_cls=VectorStoreIndex,
)
obj_retriever = obj_index.as_retriever(similarity_top_k=3)
agent_worker = FunctionCallingAgentWorker.from_tools(
tool_retriever=obj_retriever,
llm=vertex_gemini,
system_prompt="""
You are an agent designed to answer queries over a set of given papers.
Please use the tools provided to answer a question as possible. Do not rely on prior knowledge. Summarize your answer\
""",
verbose=True,
)
agent = AgentRunner(agent_worker)In this example,
all_tools is a list of all the tools you want to use. If you want to add more tools, you can modify
all_tools and recreate
obj_index,
obj_retriever, and
agent_worker.
Please note that this is a workaround and may not be the best practice. It's always recommended to initialize the
FunctionCallingAgentWorker with all the tools you plan to use.
