Use case: user-in-the-loop agent

deepset-ai / haystack

:mag: LLM orchestration framework to build customizable, production-ready LLM applications. Connect components (models, vector DBs, file converters) to pipelines or agents that can interact with your data. With advanced retrieval methods, it's best suited for building RAG, question answering, semantic search or conversational agent chatbots.

https://haystack.deepset.ai

Apache License 2.0

14.47k stars 1.7k forks source link

Use case: user-in-the-loop agent #7737

Closed mrm1001 closed 1 week ago

mrm1001 commented 1 month ago

There are existing agentic use-cases where it's important to have users accept or reject different actions before they are taken. For example, a travel bot that asks user confirmation before booking flights. These agents might be used in a multi-agent architecture.

### Tasks
- [ ] https://github.com/deepset-ai/haystack/issues/7738

mrm1001 commented 1 week ago

We will revisit the colab: https://colab.research.google.com/drive/1LY5-OcMLU4luHdNCBODpH8s53Sy0Z73-?usp=sharing if we create a new way to track chat message history.