Skip to content

fix(deps): bump the all-dependencies group across 1 directory with 11 updates #822

fix(deps): bump the all-dependencies group across 1 directory with 11 updates

fix(deps): bump the all-dependencies group across 1 directory with 11 updates #822

Workflow file for this run

name: Claude PR Review
on:
pull_request:
types: [opened, reopened, ready_for_review]
issue_comment:
types: [created]
pull_request_review_comment:
types: [created]
permissions:
contents: read
pull-requests: write
issues: write
actions: read
jobs:
claude-review:
runs-on: ubuntu-latest
timeout-minutes: 15
if: >
(github.event_name == 'pull_request' &&
github.event.pull_request.draft == false) ||
(github.event_name == 'issue_comment' &&
github.event.issue.pull_request &&
contains(github.event.comment.body, '@claude') &&
contains(fromJSON('["MEMBER", "OWNER", "COLLABORATOR"]'), github.event.comment.author_association)) ||
(github.event_name == 'pull_request_review_comment' &&
contains(github.event.comment.body, '@claude') &&
contains(fromJSON('["MEMBER", "OWNER", "COLLABORATOR"]'), github.event.comment.author_association))
steps:
- name: Checkout repository
uses: actions/checkout@v6
with:
fetch-depth: 0
- name: Generate GitHub App Token
id: generate-token
uses: actions/create-github-app-token@v3
with:
app-id: ${{ secrets.CLAUDE_APP_ID }}
private-key: ${{ secrets.CLAUDE_APP_PRIVATE_KEY }}
- name: Run Claude review
uses: anthropics/claude-code-action@v1
with:
anthropic_api_key: ${{ secrets.ANTHROPIC_API_KEY }}
github_token: ${{ steps.generate-token.outputs.token }}
prompt: |
REPO: ${{ github.repository }}
PR NUMBER: ${{ github.event.pull_request.number || github.event.issue.number }}
You are reviewing a pull request for the Ably CLI (@ably/cli), an oclif-based TypeScript CLI.
## Philosophy
You are a helpful second pair of eyes, NOT a gatekeeper. Focus on things that
actually matter — bugs, security issues, architectural problems, and correctness.
Do NOT nitpick style, formatting, naming, or minor convention deviations that don't
affect correctness. Developers read every comment you post, so make each one count.
## Setup
1. Read .claude/CLAUDE.md for project standards.
2. Read .claude/skills/ably-review/SKILL.md for the review checklist (use as a reference, not a line-by-line audit).
3. Read .claude/skills/ably-new-command/SKILL.md for command conventions (only if command files changed).
4. Check which files were changed in this PR.
## What to flag (significant concerns only)
- **Bugs**: Logic errors, race conditions, missing error handling that would crash
- **Security**: Secrets in code, injection vulnerabilities, unsafe input handling
- **Architectural mistakes**: Wrong base class, missing cleanup/teardown, broken JSON output contract
- **Correctness**: this.error() instead of this.fail() (breaks JSON mode), unguarded human output in JSON path, missing test coverage for new commands
- **Breaking changes**: Behavioral changes the PR author may not have intended
## What NOT to flag
- Style issues (eslint handles this)
- Minor naming preferences
- Missing optional improvements (extra tests, additional error messages, docs)
- Convention deviations that don't affect correctness (e.g., describe block naming)
- Anything in unchanged code
- Suggestions beyond the PR's scope
## Tone
Be direct and constructive. When flagging an issue, explain WHY it matters — what
breaks, what's the risk, what's the user impact. Don't just cite a rule.
Always post a summary comment when the review is complete, even if you have no issues to raise.
claude_args: |
--max-turns 50
--model claude-sonnet-4-6
--allowedTools "Bash(gh pr comment:*),Bash(gh pr diff:*),Bash(gh pr view:*),Bash(gh pr review:*),Bash(gh api:*),Read,Glob,Grep"