agent-service-toolkit
agent-service-toolkit copied to clipboard
JoshuaC215
Full toolkit for running an AI agent service built with LangGraph, FastAPI and Streamlit
The current model accepts any string as input. If an incorrect model name is entered or a model is provided without an API key, a 500 error will occur. ...
Would be interesting to add loading the conversation history if re-using a thread ID from a previous session. It would basically require adding a service endpoint to get the history...
We were having some [debate](https://github.com/JoshuaC215/agent-service-toolkit/pull/42#issuecomment-2395542834) about whether / how to clean up the ChatMessage class in `schema/` which is a thin wrapper around LangChain BaseMessage, which is starting to add...
PR for #38 I chose the safety check as example implementation. I have a few points I would like improve here, but let's first see if the general idea is...
I have been using the toolkit as a template to build a RAG-based agent. This agent performs 2-3 background tasks before generating a visible response for the user. Background tasks...
First version to add Human-in-the-loop. For stream only. I've created a new class InterruptMessage because it seems appropriate to treat it as a completely different entity from chat message, although...
Use case is I have tool calls that can take a few minutes and have a few large steps - think for example scraping a web site with multiple pages...
In case of, for example, an agent with a tool call that is long running (e.g. more than 60 seconds), sometimes we hit network idle timeouts and it would be...
https://github.com/JoshuaC215/agent-service-toolkit/blame/d4655414fbe10a242c6ebe95dccfed6682dd16b7/src/agents/llama_guard.py#L65
Metadata
Owner
Metadata
Full toolkit for running an AI agent service built with LangGraph, FastAPI and Streamlit