You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Is your feature request related to a problem? Please describe.
The current tool can benefit from having a pre-built embedding database of devcontainer.json files from the Microsoft VSCode Dev Containers repository. This can help improve context accuracy and speed up the generation process by using predefined embeddings.
Describe the solution you'd like
Crawl the Microsoft VSCode Dev Containers repository at https://github.com/microsoft/vscode-dev-containers/tree/main/containers.
Extract the content from the README.md files and the devcontainer.json files within each sub-folder.
Use the devcontainer schema to guide the extraction process.
Create embeddings for this content and store them in the SQLite database used by the tool.
Describe alternatives you've considered
Continuously fetching content from the MSFT repository on-demand, but this would be less efficient.
Not using predefined embeddings, which would slow down the generation process and reduce context accuracy for similar projects.
Additional context
Use the devcontainer.json schema to ensure accurate extraction and structuring of the content.
Ensure that the strategy for picking the best container image is efficient, leveraging the README.md content to infer the most appropriate image.
Steps:
Fetch and parse the README.md and devcontainer.json files from the specified Microsoft repository.
Generate embeddings for the content using the configured embedding model.
Store the resulting embeddings in the existing SQLite database inside the data/ directory.
Update the main.py logic to utilize these pre-built embeddings, improving the efficiency and accuracy of the devcontainer.json generation process.
The text was updated successfully, but these errors were encountered:
If no one is assigned to the issue, feel free to tackle it, without confirmation from us, after registering your attempt. In the event that multiple PRs are made from different people, we will generally accept those with the cleanest code.
Please respect others by working on PRs that you are allowed to submit attempts to.
e.g. If you reached the limit of active attempts, please wait for the ability to do so before submitting a new PR.
If you can not submit an attempt, you will not receive your payout.
Thank you for contributing to daytonaio/devcontainer-generator!
Is your feature request related to a problem? Please describe.
The current tool can benefit from having a pre-built embedding database of
devcontainer.json
files from the Microsoft VSCode Dev Containers repository. This can help improve context accuracy and speed up the generation process by using predefined embeddings.Describe the solution you'd like
https://github.com/microsoft/vscode-dev-containers/tree/main/containers
.README.md
files and thedevcontainer.json
files within each sub-folder.Describe alternatives you've considered
Additional context
devcontainer.json
schema to ensure accurate extraction and structuring of the content.README.md
content to infer the most appropriate image.Steps:
README.md
anddevcontainer.json
files from the specified Microsoft repository.data/
directory.main.py
logic to utilize these pre-built embeddings, improving the efficiency and accuracy of thedevcontainer.json
generation process.The text was updated successfully, but these errors were encountered: