Skip to content

Bump NCodeGroup/code-review from 1.1.2 to 1.4.0 in the ncode-review group - #14

Merged
github-actions[bot] merged 1 commit into
masterfrom
dependabot/github_actions/ncode-review-d70b02f7be
Nov 18, 2026
Merged

github-actions[bot] merged 1 commit into
masterfrom
dependabot/github_actions/ncode-review-d70b02f7be

Conversation

@dependabot

@dependabot dependabot Bot commented on behalf of github Nov 18, 2026 •

Copy link
Copy Markdown
Contributor

Bumps the ncode-review group with 1 update: NCodeGroup/code-review.

Updates NCodeGroup/code-review from 1.1.2 to 1.4.0

Release notes

Sourced from NCodeGroup/code-review's releases.

v1.4.0

Added

  • The approve input (off by default) submits a review whose verdict is Approve, with the head unchanged and no earlier should-fix thread open, as an approving review, so a ruleset can require one approval and Scorecard's Code-Review check counts it. A review with only nits approves too, when every reviewer ran in full and verification lowered or refuted nothing. Any other review stays a comment and dismisses the bot's earlier approval, and the summary says "Approved" or "Not approved" with the reason (#203, ADR-1009).

v1.3.1

Fixed

  • A review whose check run would list more than GitHub's 65,535 characters publishes: the check run's text shows as many findings, already raised, tracked, upstream, refuted and dropped entries as fit and says how many more each section has, or that all of a section's entries are in the run's trace when none of them fits, and its summary is cut to fit, so the review's threads and body are posted instead of the whole publish failing with HTTP 422 (#201).

v1.3.0

Added

  • Synthesis merges one problem found in several files, such as one rule broken the same way in three of them, into one finding: its one thread, the review's body and the check run list the other places under "Also at", so the author answers it once. A reviewer's finding in another file counts as merged only when the finding names its place, within three lines; any other is posted on its own (#193, ADR-1008).

Fixed

  • A re-review whose changed hunks are more than one part of a first review holds (160,000 characters) is a full review, in parts and under the run's cost ceiling, and the check run names the files it couldn't fit, instead of one reviewer call that stops at its $1 budget and reviews nothing (#194, ADR-1007).
  • A re-review that raises an earlier finding again in new words lists it under "Already raised", with a link to the earlier thread, instead of posting it again: its reviewer names the earlier finding it repeats, and the re-review leaves it out of its verdict (#192, ADR-1008).
  • A re-review whose verdict is Comment, because an earlier finding is still waiting or the re-review was incomplete, keeps it when the publish step lists a finding that repeats a dismissed thread as already raised, instead of turning to Approve.

v1.2.0

Added

  • A /review comment from someone with write access that starts with /review and then a space or the end of the comment but isn't a command, such as /review please, gets a 😕 reaction within seconds, so its author can tell that no review started; the run's log lists the commands. The reaction uses a token of its own that holds only the pull requests permission, minted before setup-node and revoked before the review step, and never in the review step's environment (#174, ADR-1004).
  • npm run train learns from the review bot's own findings that the maintainer answered, one source per pull request with the replies and whether each finding was fixed or declined: a decline with a lasting reason teaches reviewers what not to report, and a fix the rule it enforced. Only the maintainer's replies count (ADR-1005).

Changed

  • The repository follows project-template v0.68.4 (from v0.59.0): the Commit attribution check and the commit-msg hook catch attribution from any AI agent or model, and the check's remedy for an attribution line in a commit message runs the hook, as CONTRIBUTING.md now says; the workflow commands remove an agent footer a session appends to what they post, except on a /review comment, which /run-issues and /run-feedback leave as posted because NCode Review reads the command from its first line; the commands that call gh use its REST forms in a session that refuses GitHub's GraphQL API, such as a Claude Code cloud session; /address-feedback and /run-feedback also run in Copilot Chat and Copilot CLI and finish a run that stopped, also one the other tool started; .template/Update-Project.ps1 stages only the files an update changes; and AGENTS.md's Landing section sets pre-release: tracked, the template's new key, as it is for a project with releases.
  • The repository follows project-template v0.75.1 (from v0.68.4): Claude Code sessions run Bash and PowerShell commands without asking, except git push and a command that mentions .env or secrets, and a Claude Code cloud session installs pwsh when its container lacks it; /ship-plan can mark follow-ups to act on next, which /finish-plan then names as /plan-issue runs; the Copilot setup workflow runs again when .node-version changes; and .template/Initialize-Nvs.ps1 sets nvs up once per PC, as CONTRIBUTING.md says. The template's tag-release.yml publishes nothing here, because release stays none: releases stay manual through release.yml, the Release (manual) workflow in the Actions tab.
  • The repository follows project-template v0.83.0 (from v0.75.1): nothing checks commits or pull requests for AI agent attribution, so the Commit attribution CI job, .template/Test-CommitAttribution.ps1 and the .githooks/commit-msg hook are removed, and CONTRIBUTING.md asks contributors to avoid it (the template's ADR-0018); CI doesn't run again when a pull request's title or description is edited; a Claude Code cloud session runs each .claude/hooks/ensure-*.sh at start, which installs pwsh and the .node-version Node.js; /run-issues and /ship-plan can fix a plan's follow-ups in its pull request before merge; and plans can land through a branch a session's host made for it. The prelude is the template's .template/Use-Node.ps1, which accepts the .node-version Node.js already on PATH where nvs isn't installed, and tools/Use-Node.ps1 is removed (ADR-1003).
  • The repository follows project-template v0.84.1 (from v0.83.0): .template/Update-Project.ps1 stages every change outside the conflicting files, deletions included, when it stops on a conflict, and leaves a catch-all text rule such as * text eol=lf out of the .gitattributes lines an adopted project kept; /plan-template-update leaves a deletion an older target's script left unstaged unmarked, so its commit note stages it (#181).
  • The repository follows project-template v0.84.3 (from v0.84.1): the workflow commands record a Claude Code cloud session's push branch without remarking on it, and /ship-plan names the branch it pushes to in the summary it shows before it pushes; Claude Code's planning commands enable .githooks and record the session's push branch from the main checkout before they enter the plan's worktree, and /run-issues and /address-feedback check the hooks before they create theirs; and the ask rules in .claude/settings.json match a command that mentions the root secrets/ folder, not any text containing secrets, so a source folder named Secrets no longer asks for approval.
  • The repository follows project-template v0.86.1 (from v0.84.3): Copilot Chat, Copilot CLI and the VS Code agents keep each plan in its own git worktree under .claude/worktrees/, as Claude Code does, as CONTRIBUTING.md says; /run-issues and /address-feedback in Copilot work in worktrees of their own and never switch the main checkout, and /run-issues in every tool runs the next entry ahead in its own worktree while a draft pull request waits, then lands the entries one at a time, in queue order; /analyze-architecture and /analyze-decision work in a worktree of their own, where ANALYSIS.md stays untracked and a decision record lands with the plan made there, as docs/decisions/README.md says, and /continue-plan lists and /finish-plan removes such worktrees; and /finish-plan fast-forwards the default branch without switching a checkout that holds other work.
  • The reviewers settle in the checkout what they would otherwise ask, and don't report a production rule in test code or a C# rule in a script, a rule a line broke before a change that only renames or renumbers on it, compatibility or migration notes in a repository whose conventions say it is pre-release, another repository's process rules, duplication between places that can't share code, a parameter the conventions require as unused, or a date against today's. The rule check reports only what the change writes, synthesis merges one problem raised at several places in one file into one finding, tests names the search behind a missing test, and data-model and api-compat expect no earlier data or callers where nothing is released (ADR-1005).
  • The learned guidance is regenerated from NCodeGroup/identity, identity-platform and project-template: their decision records, instruction files and review prompts, and the bot's findings the maintainer answered on their pull requests. It holds 1,234 rules, 157 of them on what not to report; production and test rules name their scope; and the rules of decision records that no longer exist are gone (#190).
  • A release is made by merging: a plan whose change reaches consumers carries a Release: line, its last step dates the changelog section, and merging its pull request tags and publishes the release, so nobody runs release.yml, which stays as the fallback. This repository's own pin of NCode Review is set to follow a release within a weekday: its github-actions Dependabot entry runs daily and its cooldown excludes the action, which Dependabot's documentation doesn't say applies to github-actions, so without it the bump waits the 3 days (#196, ADR-1006).

Fixed

  • A training run deletes the rules it extracted from a source the repository no longer has, such as a renumbered decision record, so they no longer reach the guidance; --repo finds the repository's extraction folder ignoring case; and a rule learned from an instruction file for production or test code says which, so reviewers no longer apply production rules to test projects (ADR-1005).
  • Training reads a repository's pull request comments one page at a time, by number, so it runs behind a proxy that refuses GitHub's next-page links, such as a Claude Code cloud session's.
Changelog

Sourced from NCodeGroup/code-review's changelog.

Changelog

All notable changes to this project will be documented in this file.

The format is based on Keep a Changelog, and this project adheres to Semantic Versioning.

[Unreleased]

Changed

  • The template's node_lint answer is on, matching the project's Biome lint: template updates render CI with its npm run lint step instead of treating it as the project's own, VS Code recommends the Biome extension, and docs/best-practices-criteria.md counts coding_standards_enforced, warnings and warnings_fixed as met (#219).
  • CI fails when the tests cover less than 90% of the statements or 80% of the branches: npm test writes lcov.info with Node's test coverage, CI's Coverage step and the Landing gate in AGENTS.md check it with ./.template/Test-Coverage.ps1, and tag-release.yml, which runs CI on a release's commit, and the Release (manual) workflow, which runs the lint and the coverage check too, tag a release only once it passes; CONTRIBUTING.md, docs/assurance-case.md and docs/best-practices-criteria.md say so (#217).
  • Each source file starts with a Copyright © NCode Group line and an SPDX-License-Identifier: Apache-2.0 line, as the template's file_headers answer sets: CI, the Release (manual) workflow and the Landing gate in AGENTS.md run ./.template/Test-FileHeaders.ps1, which fails on a source file without them, and -Fix adds them; AGENTS.md, CONTRIBUTING.md, README.md and docs/best-practices-criteria.md say so, which meets the OpenSSF Best Practices gold criteria copyright_per_file and license_per_file (#218).
  • The project follows project-template v0.103.1: the workflow commands, prompts and skills, the file-header check (Astro components, stylesheets and hand-written declaration files), and the template's decision records are refreshed; the worktree setup runs the install in the main checkout when it has none, and /ship-plan and /run-feedback close the open follow-up issues a plan's work completes.

Fixed

  • The Release (manual) workflow, release.yml, signs the tag it makes with gitsign, as tag-release.yml does, so gitsign verify-tag checks a fallback release's tag too, and SECURITY.md gives one command that verifies a tag from either workflow; its checks run in a job that can only read, and the job that tags, signs and publishes runs no npm (#216).

[1.4.2] - 2026-11-16

Added

  • The repository is licensed under the Apache License 2.0, as the other NCodeGroup repositories are: LICENSE.txt holds the license and COPYRIGHT.txt the copyright notice.
  • .git-blame-ignore-revs lists the commits of the one-time Biome reformat, so GitHub's blame view and git blame (with git config blame.ignoreRevsFile .git-blame-ignore-revs) show who last changed a line instead of the reformat (#212, ADR-1010).

Changed

  • GOVERNANCE.md, ROADMAP.md, docs/architecture.md and docs/assurance-case.md describe code-review itself: its maintainer, its plans for the next twelve months, its components and design rules, and its security requirements, threat model and the weaknesses it counters (#206).
  • README.md shows the OpenSSF Best Practices badge (project 16752, passing) and says how to verify a release with gh release verify, and docs/best-practices-badge.md records where the silver level stands.
  • CI runs npm run lint, which fails on TypeScript that Biome's formatter or linter doesn't accept as biome.json sets them, and npm run format applies them; the code is reformatted once to match, and AGENTS.md, CONTRIBUTING.md and docs/best-practices-badge.md say so (#210, ADR-1010).
  • The repository follows project-template v0.99.0 (from v0.96.0): it gains the /best-practices command, which checks the OpenSSF Best Practices answers against the repository, and docs/best-practices-criteria.md, which maps every criterion to its evidence; tag-release.yml runs CI on a release's commit before it tags it and signs the tag with gitsign, which SECURITY.md and README.md say how to check; CONTRIBUTING.md has a Review Requirements section, SECURITY.md promises an acknowledgement within 14 days, and docs/assurance-case.md has a Security Reviews section.

Fixed

  • The action loads again: action.yml, from 1.4.0 on, wrote the approve input's description as a plain YAML scalar holding : , which GitHub's runner can't parse, so every job that used the action failed in "Set up job". A test parses action.yml as YAML (#221).

[1.4.1] - 2026-11-15

Changed

  • The repository follows project-template v0.96.0 (from v0.86.1): it gains CODE_OF_CONDUCT.md, GOVERNANCE.md, ROADMAP.md, docs/architecture.md, docs/assurance-case.md and docs/best-practices-badge.md, the documents the OpenSSF Best Practices silver level asks for, an OpenSSF Scorecard workflow (.github/workflows/scorecard.yml), read-only top-level permissions in .github/workflows/copilot-setup-steps.yml, and the template's newer review commands, which judge every NCode Review finding whatever its severity; AGENTS.md, CONTRIBUTING.md, README.md and docs/README.md describe them.

[1.4.0] - 2026-11-15

Added

  • The approve input (off by default) submits a review whose verdict is Approve, with the head unchanged and no earlier should-fix thread open, as an approving review, so a ruleset can require one approval and Scorecard's Code-Review check counts it. A review with only nits approves too, when every reviewer ran in full and verification lowered or refuted nothing. Any other review stays a comment and dismisses the bot's earlier approval, and the summary says "Approved" or "Not approved" with the reason (#203, ADR-1009).

[1.3.1] - 2026-11-14

... (truncated)

Commits
  • d123366 Merge pull request #204 from NCodeGroup/worktree-issue-203
  • 923ffc1 Plan step 11: Fix the docs wrapping and wording (review round 1)
  • 1376ce0 Plan step 10: Test the untested approval paths (review round 1)
  • 14a844d Plan step 9: Fix the approval decision and its wiring (review round 1)
  • 4d29142 Plan step 8: Describe nit-only approval in the docs and changelog
  • a6e29c2 Plan step 7: Approve a review whose only findings are nits
  • 249ac07 Plan step 6: Write the changelog entry and run the gate
  • a9e11e1 Plan step 5: Record the decision and document the input
  • 5266786 Plan step 4: Wire the approval into the publish step and action.yml
  • 35fbe11 Plan step 3: Add the approval line to the summary comment
  • Additional commits viewable in compare view

Dependabot compatibility score

Dependabot will resolve any conflicts with this PR as long as you don't alter it yourself. You can also trigger a rebase manually by commenting @dependabot rebase.


Dependabot commands and options

You can trigger Dependabot actions by commenting on this PR:

  • @dependabot rebase will rebase this PR
  • @dependabot recreate will recreate this PR, overwriting any edits that have been made to it
  • @dependabot show <dependency name> ignore conditions will show all of the ignore conditions of the specified dependency
  • @dependabot ignore <dependency name> major version will close this group update PR and stop Dependabot creating any more for the specific dependency's major version (unless you unignore this specific dependency's major version or upgrade to it yourself)
  • @dependabot ignore <dependency name> minor version will close this group update PR and stop Dependabot creating any more for the specific dependency's minor version (unless you unignore this specific dependency's minor version or upgrade to it yourself)
  • @dependabot ignore <dependency name> will close this group update PR and stop Dependabot creating any more for the specific dependency (unless you unignore this specific dependency or upgrade to it yourself)
  • @dependabot unignore <dependency name> will remove all of the ignore conditions of the specified dependency
  • @dependabot unignore <dependency name> <ignore condition> will remove the ignore condition of the specified dependency and ignore conditions

This change is Reviewable

Bumps the ncode-review group with 1 update: [NCodeGroup/code-review](https://github.com/ncodegroup/code-review).


Updates `NCodeGroup/code-review` from 1.1.2 to 1.4.0
- [Release notes](https://github.com/ncodegroup/code-review/releases)
- [Changelog](https://github.com/NCodeGroup/code-review/blob/main/CHANGELOG.md)
- [Commits](NCodeGroup/code-review@564bb16...d123366)

---
updated-dependencies:
- dependency-name: NCodeGroup/code-review
  dependency-version: 1.4.0
  dependency-type: direct:production
  update-type: version-update:semver-minor
  dependency-group: ncode-review
...

Signed-off-by: dependabot[bot] <support@github.com>
@dependabot dependabot Bot added dependencies Pull requests that update a dependency file github_actions Pull requests that update GitHub Actions code labels Nov 18, 2026
@github-actions
github-actions Bot enabled auto-merge November 18, 2026 23:55
@github-actions
github-actions Bot merged commit 05bbbfc into master Nov 18, 2026
7 of 8 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

dependencies Pull requests that update a dependency file github_actions Pull requests that update GitHub Actions code

Projects

None yet

Development

Successfully merging this pull request may close these issues.

0 participants