Skip to content

docs(plans): add setup-cli code quality assessment - #161

Open
bens-agent wants to merge 1 commit into
mainfrom
docs/code-quality-assessment
Open

bens-agent wants to merge 1 commit into
mainfrom
docs/code-quality-assessment

Conversation

@bens-agent

@bens-agent bens-agent commented Nov 10, 2026 •

Copy link
Copy Markdown
Collaborator

Requested by bens-agent · project thread

Summary

This PR adds a written assessment of code quality gaps in this repository, at plans/backlog/20261110-code-quality-assessment.md. It changes no code. The doc has 16 findings in three priority tiers, each with what's wrong, why it matters, a suggested fix and a rough effort. It ends with a suggested order for the work.

P1 (fix first):

  • The last release failed. v0.3.0 was tagged twice, and the Release workflow failed both times because package.json still says 0.2.4. So @v0 still points at v0.2.4 from May, which installs CLI v0.10.1 as "latest", while main says v0.12.4.
  • "latest" is a constant in src/versions.ts, so every CLI release needs a PR and a release here before users see it.
  • The downloaded CLI archive is not checked against the cli_<version>_checksums.txt the CLI release already publishes.
  • A bad or unknown version (v0.13, v1, a typo) fails with a bare Unexpected HTTP response: 404.

P2: two conflicting release processes with a manual package.json bump, macOS runners rejected although the CLI ships Darwin builds, no tool cache, a hand-rebuilt dist/ on every Dependabot PR, tests that mock the whole pipeline and a CI matrix with no arm64 or partial-version job, and a stale license cache that is never checked.

P3: lint without type information (it would catch an await on a plain string in main.ts), unused dependencies and template leftovers, a devcontainer and local lint that don't match CI, a README.md with no usage, workflow hygiene, and repository hygiene (stale branches, one code owner, no AGENTS.md).

How it was checked

  • The whole repository was read at 04ffda4.
  • With Node 24.4.0, npm ci, format check, ESLint and tests all pass (44 tests, 100% coverage), and npm run package reproduces dist/ with no diff.
  • Release history and the failed Release runs were read from the GitHub API.
  • The CLI's checksums file and macOS archives were confirmed by downloading them.
  • The doc passes markdownlint, Prettier and textlint terminology locally.

Generated by Claude Code


Autofix with [code]smith
Open in [code]smith · Need help on this PR? Tag @codesmith-bot with what you need.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01XHScKaZf6HW9aiXQnDjCAM
@bens-agent
bens-agent requested a review from ben-miru as a code owner November 10, 2026 23:33
@chatgpt-codex-connector

chatgpt-codex-connector Bot commented Nov 10, 2026 •

Copy link
Copy Markdown
Codex Review Summary: ✅ Completed

This comment shows the latest Codex review activity on this pull request.

Review Status Commit Review trigger
📝 Code Review ✅ Completed 2026-11-10T23:34:11.094899Z ab5f39d PR opened
🔒 Security Review ✅ Completed 2026-11-10T23:35:26.951867Z ab5f39d PR opened
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" or "@codex security review".

Codex reacts with 👀 while any review is running, comments if it has suggestions, and reacts with 👍 once all reviews finish with no findings.

@github-actions

Copy link
Copy Markdown

AI review: approve for ab5f39d

I'm approving because the PR only adds a working-notes file under plans/, which the review rules exclude from findings. It changes no code, workflow or config. Neither reviewer reported anything, and there are no open threads or earlier blocking points to re-check.

Review run

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