Visitar URL original
fix(connect): label the sign-in code and say where the signup email came from by marckong · Pull Request #485 · Kong/volcano-cli · GitHub
Skip to content

fix(connect): label the sign-in code and say where the signup email came from - #485

Open
marckong wants to merge 4 commits into
mainfrom
marckong/connect-output-clarity
Open

marckong wants to merge 4 commits into
mainfrom
marckong/connect-output-clarity

Conversation

@marckong

@marckong marckong commented Nov 12, 2026 •

Copy link
Copy Markdown
Contributor

Tracking

Why

  • Browser sign-in printed Code: ABCD-EFGH with no label. A coding agent reading that output can take an unlabelled token for an API key and stop (VOL-1535).
  • volcano connect --signup silently used the global git email for signup. A signed-out user saw their address appear with no explanation (VOL-1652).

Summary

  • Browser login and signup now print One-time sign-in code (not a secret): ABCD-EFGH instead of Code: ABCD-EFGH. This is the OAuth user code the approval page shows, not the device code or a token.
  • When connect --signup takes the address from git config --global user.email (no --email), it prints a note: "The signup link includes , your git config --global user.email. You can sign up with any address on the signup page." No note when --email is given. The address shows up in the printed signup link, which is where the user sees it.
  • The note goes to connect's log writer, which is stderr under --json, so JSON stdout is unchanged. The --json result shape is unchanged.
  • docs/connect.md describes the labelled code, how it differs from the dashboard prompt's --device-code (which is a secret), and the git-email note.
  • deferred: whether the web signup page should prefill and explain the address (it ignores the email parameter today).
  • Out of scope: the dashboard setup prompt's inline --device-code (VOL-1535's web part, another PR).

Verification

  • make check (openapi-generated-check, release-workflow-check, golangci-lint 0 issues, make test all packages ok)
  • go build ./..., make build FIRST_PARTY_DEVICE_CLIENT_ID=volcano-cli, ./volcano --version
  • Shared docs lint from Kong/volcano-docs (scripts/lint-docs.mjs docs): OK
  • New test TestConnectSignupSaysTheEmailCameFromGit (text and --json: note on the log stream only); TestConnectSignupUsesEmailFlagOverARelayGitEmail now asserts no note with --email; auth_test asserts the new label.
  • The volcano-web signup page at main (3c319ac16) reads only next and ref, not the email the CLI sends, so the note does not claim the page prefills it or that the account uses it.
  • Independent pr-review and team-review; their P2s (note wording, user code vs dashboard device code) are fixed in the last commit.

@marckong
marckong requested a review from a team as a code owner November 12, 2026 23:56
Copilot AI balanced review requested due to automatic review settings November 12, 2026 23:56

Copilot AI left a comment

Copy link
Copy Markdown

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Copilot was unable to review this pull request because the user who requested the review has insufficient remaining quota. Try requesting a review at a lower effort level.

@github-actions

Copy link
Copy Markdown

AI review: no P0-P2 findings

Reviewed a68b308. Run log

Relabels the browser sign-in user code as a "One-time sign-in code (not a secret)", adds a note when connect --signup takes the signup address from git config --global user.email (sent to the log writer, so stderr under --json), and updates docs/connect.md to match. Low risk: output and docs wording only, with tests covering text and JSON modes.

Not checked: The AI review read the diff, the code, and the PR's text; it ran no tests or builds and read nothing outside this repository.

  • Whether the volcano-web signup page still ignores the email query parameter, which the note's wording assumes.
  • Confirm Volcano Web’s signup page permits choosing a different email address, as the new CLI note states.

Human review: not needed

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants