Support fully autonomous mode for agent #1798
Labels
agent framework
strategies for prompting, agent, etc
architecture
related to architecture, including frontend and backend
enhancement
New feature or request
frontend
issues related to frontend
severity:low
Minor issues, code cleanup, etc
What problem or use case are you trying to solve?
Discussion starts here: #1755 (comment)
Originally posted by @li-boxuan in #1755 (comment)
Describe the UX of the solution you'd like
Maybe add a checkbox in the front-end that allows the (frontend? backend?) system to automatically provide the prompt ("Please continue working on the task on whatever approach you think is suitable...") whenever the model asks for human inputs.
The user may want it to run autonomously for a while, then uncheck the box; they expect the model will go back to the mode that requests human input.
When autonomous mode finishes, it may or may not solve the user's task (but will emit the
AgentFinish
action). We need to consider how to handle this - should we allow users to follow up given the same context?cc @li-boxuan
Do you have thoughts on the technical implementation?
Describe alternatives you've considered
Additional context
The text was updated successfully, but these errors were encountered: