docs: clarify runtime provenance boundaries - #226
Conversation
Codex Review SummaryThis comment shows the latest Codex review activity on this pull request.
ℹ️ About Codex in GitHubYour team has set up Codex to review pull requests in this repo. Reviews are triggered when you
Codex reacts with 👀 while any review is running, comments if it has suggestions, and reacts with 👍 once all reviews finish with no findings. |
![chatgpt-codex-connector[bot]](https://avatars.githubusercontent.com/in/1144995?s=60&v=4){kind=small}
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: 3823c70c05
ℹ️ About Codex in GitHub
Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you
- Open a pull request for review
- Mark a draft as ready
- Comment "@codex review".
If Codex has suggestions, it will comment; otherwise it will react with 👍.
Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".
| Each runtime edge has a required `evidenceRefs` array. The current Docker and | ||
| Systemd slices emit bounded, versioned records alongside the edge during | ||
| Each runtime edge has a required `evidenceRefs` array. The current Docker, | ||
| Systemd, and npm slices emit bounded, versioned records alongside the edge during |
There was a problem hiding this comment.
For npm dependencies, the collector does not emit a RuntimeEvidenceRef alongside the edge: providers/npm.rs creates an edge with evidence_refs: Vec::new() and a private marker, and bind_npm_evidence() in cache_refresh.rs creates the evidence only after validating the project_npm slot revision and successful collection timestamp. Including npm in this collector-to-evidence description misdocuments the fail-closed trust boundary; document the intermediate scheduler-slot binding step instead.
Useful? React with 👍 / 👎.
1 participant
Summary
Checks
git diff --checkNo code, contracts, or generated artifacts changed.