New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
[Security solution] AI Assistant, replace LLM
with SimpleChatModel
#182041
base: main
Are you sure you want to change the base?
Conversation
memory, | ||
verbose: false, | ||
returnIntermediateSteps: false, | ||
handleParsingErrors: 'Try again, paying close attention to the allowed tool input', |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
handleParsingErrors
can also be a function, so I put logs in here while developing. It was not hit once I added the agentArgs
. However when I was hitting it, it did help to get the agent back on track so I think I should leave it. Here is an example of a run where it was hit: https://smith.langchain.com/public/910b7739-cd9a-401b-9db0-fe8438ff53e5/r
Summary
LLM
model andchat-conversational-react-description
agent with theSimpleChatModel
model andstructured-chat-zero-shot-react-description
agent for non-streaming AI Assistant requests.ESQLKnowledgeBaseTool
fromDynamicTool
toDynamicStructuredTool
Here are some datasets using the updated
ESQLKnowledgeBaseTool
SimpleChatModel - Sonnet
SimpleChatModel - Azure
ChatOpenAI - Azure
Still To Do
AlertCountsTool
andOpenAndAcknowledgedAlertsTool
tools fromDynamicTool
toDynamicStructuredTool
ActionsClientLlm
withActionsClientSimpleChatModel
(post_evaluate
andpost_attack_discovery
)AttackDiscoveryTool
toDynamicStructuredTool
May do the attack discovery stuff in a separate PR depending on what @andrew-goldstein thinks.