-
Notifications
You must be signed in to change notification settings - Fork 99
feat: moving negotiation bots to arklex #158
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
base: main
Are you sure you want to change the base?
Conversation
…com/arklexai/Agent-First-Organization into ma4407/move_negotation_single_issue
yongwhan
left a comment
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.
could you put some comments (e.g., docstrings) in your code?
luyunan0404
left a comment
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.
Could you put config and prompts under - negotiation_workers folder as well? So eveything related to negotiation are in one folder.
Please add the command to run your testing and paste your test result if possible.
arklex/env/workers/negotiation_workers/negotiation_single_issue_worker.py
Outdated
Show resolved
Hide resolved
arklex/env/workers/negotiation_workers/negotiation_single_issue_worker.py
Show resolved
Hide resolved
Yes will do. @luyunan0404 where should I add the command? Should I just add it to the description for this PR? |
…der negotation_workers and removing taskplanning files
…com/arklexai/Agent-First-Organization into ma4407/move_negotation_single_issue
I've added them in the single issue and multi issue workers, let me know if you think there are any other places I should add them! |
luyunan0404
left a comment
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.
lgtm without breaking the current logic.
yongwhan
left a comment
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.
lgtm
|
make sure to update the branch; then, should be good for merging? @luyunan0404 |
yongwhan
left a comment
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.
make sure to add some unit tests
yongwhan
left a comment
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.
lgtm
Coverage reportClick to see where and how coverage changed
This report was generated by python-coverage-comment-action |
||||||||||||||||||||||||
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.
can we remove the changes here
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.
I think there was a ruff error which is why when I ran pre-commit it fixed this file, but will double check
Summary
We implemented two negotiation workers: a Single Issue Worker for straightforward price negotiations and a Multi Issue Worker for handling complex negotiations involving multiple aspects (financing, warranty, color, price). The Multi Issue Worker includes advanced features like KDE (Kernel Density Estimation) for analyzing offer patterns, utility calculations for both buyer and seller, and a monitoring system to track resolved issues. Both workers use MessageState for conversation tracking and share a consistent slot structure with proper metadata. Recent improvements focused on streamlining the code by removing redundant history tracking, fixing conversation flow issues (like double ice breakers), implementing proper JSON handling for complex data in slots, and ensuring clean integration with the orchestrator through a single-node task graph configuration. The result is a robust negotiation system that can handle both simple and complex negotiations while maintaining natural conversation flow and effective state management.
Description
This is needed to help scale the multi and single issue bots in the future.
Tests
I ensured that the run.py command works as expected and all parts of the bots are executing properly.
Here are the commands I used:
python run.py --input-dir ./examples/negotiation_agent_multi_issue (for the multi issue bot)
python run.py --input-dir ./examples/negotiation_agent (for the single issue bot)
Here is a screenshot of the multi issue conversation:
(insert screenshot here)
Reviewers