-
Notifications
You must be signed in to change notification settings - Fork 200
Integrate long-term memory for build error handling #1160
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
/gcbrun exp -n wq -ag |
No need to worry about lint/format CI failure at the moment given the main goal is iterating on the best approach. |
/gcbrun exp -n wq -ag |
/gcbrun exp -n wq -ag -rc |
/gcbrun exp -n wq -ag -rag |
/gcbrun exp -n wq-rag -ag -- -rag |
/gcbrun exp -n wq -m vertex_ai_gemini-2-5-pro-chat -ag |
/gcbrun exp -n wqrag -m vertex_ai_gemini-2-5-pro-chat -ag -- -rag |
/gcbrun exp -n wqrag -m vertex_ai_gemini-2-5-pro-chat -ag |
This PR brings RAG-based long-term memory into the Prototyper for build error handling.
First, the Prototyper now has new prompt templates and an updated workflow so it can pull in guidance from stored error knowledge when dealing with build failures.
Second, under the
helper
folder there are two additions:error_classifier.py
– to categorize build errors.error_patterns.yaml
– a small database of regex patterns with “good” and “bad” strategies, built from past fixes.With this setup the Prototyper can match new errors against known patterns, retrieve useful hints, and reuse past reasoning to handle recurring build issues.