-
Notifications
You must be signed in to change notification settings - Fork 290
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
feat: add E2B code interpreter #106
base: main
Are you sure you want to change the base?
feat: add E2B code interpreter #106
Conversation
Hi @lorenzejay, I've just improved the text and added comments to the README to explain why we have a |
Awesome. Again, should it just automatically use the close method once it finishes the response? Don't need to keep the process running when its done right. Instead of having users set close manually. not saying remove the method just include that or even include that as a flag users can set ie: |
@lorenzejay If I understand correctly that would close the sandbox after the first time the tool is called. Since we don't know how many times the tool will be called (and it's frequently used more than once) that would cause workflows to break. |
@jamesmurdza getting close then on root init.py: |
@jamesmurdza when running the tool on my crew i get this: File ".../pypoetry/virtualenvs/crewai-Nmm_nbE4-py3.11/lib/python3.11/site-packages/e2b_code_interpreter/main.py", line 40, in init |
This is a plug-and-play alternative of the Docker code interpreter that uses E2B cloud sandboxes.
The inputs and outputs are slightly different compared to the default code interpreter:
Input: Dependencies are specified in the code via "!pip install"
Output: Output is a JSON structure separating standard output, errors, and rich data.
However these don't necessarily affect the end user as they are specified for the agent in the schema.
It also requires an E2B API key.