Conversation
Installs the LiteLLM proxy on a private, uv-managed Python runtime under /usr/local/share/litellm, linked as /usr/local/bin/litellm. The sandbox runner drives it to bridge Codex (Responses API) and Claude Code (Messages API) onto Chat Completions-only upstreams such as LithosAI. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01Mr8awhwwWhJpkEKKHkEZpq
![greptile-apps[bot]](https://avatars.githubusercontent.com/in/867647?s=60&v=4){kind=small}
The build tool is a fixed uv release fetched as the wheel PyPI publishes for the image's architecture and checked against that wheel's digest before it runs, instead of piping a mutable installer script into sh. The package default moves from latest to a reviewed 1.101.0, with latest kept as an explicit opt-in. A scenario now exercises a non-default pythonVersion and asserts the launcher runs on the requested interpreter. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01Mr8awhwwWhJpkEKKHkEZpq
This file contains hidden or 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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Summary
Adds a
litellmfeature that installs the LiteLLM proxy on a private, uv-managed Python runtime under/usr/local/share/litellm, linked as/usr/local/bin/litellm.The sandbox runner will drive this proxy to bridge Codex (Responses API only) and Claude Code (Anthropic Messages API only) onto Chat Completions-only upstreams, LithosAI first. See the companion PRs in
orianna-ai/devcontainer(adds the feature to the image) andorianna-ai/orianna(thelitellmprovider onrun_claude/run_codex).What the feature does
uvrelease (0.12.15) as the wheel PyPI publishes for the image's architecture, verifies it against that wheel's sha256 embedded in the script, and uses it only from a staging directory that is removed afterwards.pythonVersion, default 3.13) andlitellm[proxy](version, default the reviewed 1.101.0;latestis an explicit opt-in) into the shared prefix. Nothing lands in a home directory and the image's own Python is untouched.Testing
install.shin a clean Ubuntu container for the default and forpythonVersion=3.12: the launcher resolves inside the prefix, its shebang points at the private interpreter of the requested version,litellm --versionreports 1.101.0, and the proxy answers/health/livelinessabout 6s after start with a wildcard config.ubuntu(defaults; asserts the reviewed default version, prefix layout, private interpreter, no home leftovers, read-only install, liveliness),ubuntu_latest, andubuntu_python(asserts the launcher runs on the requested 3.12 runtime).pre-commitpasses (shellcheck, shfmt, JSON formatting).🤖 Generated with Claude Code
https://claude.ai/code/session_01Mr8awhwwWhJpkEKKHkEZpq