Incorporate changes from deciduus agent-zero fork [~ IDE terminal compatibility, docker quality of life, prompt optimizations, execution handling & style encoding, NVIDIA CUDA (GPU support) ~ #19] #1
+880
−180
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
added pycharm ide running compatibility
slight edit of prompt
slight revision of prompt
refined prompts for clarity and conciseness adding some new steps for logic and reducing chance of subordinate looping increased main prompt timeout to 2 min so it won't run so quick for so little value
…jects. made edits to the docker.py to include a health check (threaded), it does not impact framework flow. also added a check and reconnect mechanic for the docker due to some issues where it would disconnect, try terminal commands, and get no response (wasting api responses). the health check thread also follows up to close the docker container upon force close in IDE, which helps save computer resources. knowledge tool was updated to include a top 3~ hits for wikipedia content to build the contextual awareness. timed input was adjusted upward for testing.
…d errors (lack of gpu resources), to prevent the error in tool. docker wait time was decreased as the connection doesnt take that long print style added encoding utf-8 to prevent read errors, this resulted in many terminal tasks that were previously "erroring" due to simply taking too long or having a bad symbol to actually function better.
edited docker file and readme to allow for nvidia cuda support, giving access to gpu resources. added readme steps to enable the GPU compatibility setup
…d perms for container. added commented out code in dockerfile for trying to import torch and numpy at start, but currently having issues with it initiating (but pushing so others have it incase they want to try to fix)
- build.txt same ^ - code execution tool got elaborate updates for prompt-handling pass-back, idle timeout, comparing old and new outputs to prevent spamming user with identitical terminal commands, and a summarized output feature to prevent slamming API with too many tokens on big terminal tasks, also the terminal will retain control and the agent will not check in unless idle proc (about 30 seconds) -docker.py sleeps slightly longer anticipating some may take longer to spin up the new container. -docker compose has been added for the user to create the nvidia image - please see readme for detailed instructions -dockerfile has been adjusted, new container (old is commended out), timezone check for automatic selection on image compose. organized a bit. -main.py adjusted the commented out docker names just for clarity -adjusted rate_limiter.py to fix a bug when the agent was doing long tasks and lots of calls -added intensive README section for ubuntu and nvidia cuda runtime setup and troubleshooting. -timeout_input was reverted to original value
…plified) extensive dockerfile changes to remove venv and add custom package installs when composing docker image. (machine learning packages) adjusted the image reference comment on main for my fork readme is a bit of a mess (not gracefully formatted additions) but i added some troubleshooting info for docker/ubuntu composing, images etc.
…ide (for redundant reference) code execution tool summarizing extended for more context on longer files.
|
creating a merge commit from User: deciduus |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
I'm new here, but continuing to play around and tweak -- as per : but I'm submitting a pull request as these seem like good QoL for everyone.
deciduus-agent-zero repo
User: deciduus development pulls ~ IDE terminal compatibility, docker quality of life, prompt optimizations, execution handling & style encoding, NVIDIA CUDA (GPU support) ~ frdel#19
From User: deciduus :
I think the latest commit from a few minutes ago has several useful features that could benefit the main project.
Docker Enhancements:
Added a threaded health check in docker.py to monitor container status without impacting framework flow.
Implemented a check and reconnect mechanism to handle Docker disconnections, preventing wasted API responses.
The health check thread ensures Docker containers are properly closed upon force close in IDE, conserving computer resources.
Knowledge Tool Update:
Enhanced the tool to include the top ~3 hits from Wikipedia, improving contextual awareness.
Timed Input Adjustment:
Increased the timeout duration for testing purposes.