Skip to content

[pull] main from erikdarlingdata:main #3

[pull] main from erikdarlingdata:main

[pull] main from erikdarlingdata:main #3

Workflow file for this run

name: Claude Auto Review
# Reviews every pull request into dev or main when it is opened or updated. The review is
# advisory (it posts comments) and is deliberately NOT a required status check, so it never
# blocks a merge. It no-ops cleanly until the CLAUDE_CODE_OAUTH_TOKEN repo secret is set, and on
# fork PRs (which do not receive secrets), so neither case shows a failed check.
on:
pull_request:
types: [opened, synchronize, reopened]
branches: [dev, main]
# Cancel an in-flight review when new commits are pushed to the same PR.
concurrency:
group: claude-review-${{ github.event.pull_request.number }}
cancel-in-progress: true
jobs:
review:
runs-on: ubuntu-latest
# Reviews are advisory and typically finish in minutes; a wedged run should release
# its runner rather than sit on the default 6h timeout.
timeout-minutes: 30
permissions:
contents: read
pull-requests: write
id-token: write
env:
# Mapped to env so the step guard below can skip cleanly when the secret is absent.
CLAUDE_CODE_OAUTH_TOKEN: ${{ secrets.CLAUDE_CODE_OAUTH_TOKEN }}
steps:
- uses: actions/checkout@v7
with:
fetch-depth: 1
- name: Claude review
if: ${{ env.CLAUDE_CODE_OAUTH_TOKEN != '' }}
uses: anthropics/claude-code-action@v1
with:
claude_code_oauth_token: ${{ env.CLAUDE_CODE_OAUTH_TOKEN }}
prompt: |
REPO: ${{ github.repository }}
PR NUMBER: ${{ github.event.pull_request.number }}
Review this pull request. Follow the conventions in CONTRIBUTING.md — including
its T-SQL style section (AS on table aliases, column_name = expression aliasing,
OPTION(RECOMPILE) on collector queries). This repository ships two apps that must stay in parity — Lite
and Darling — so flag any change made to one but not its counterpart. Focus on:
- Correctness: bugs, edge cases, null/error handling at system boundaries
- Lite/Darling parity drift
- Security: input handling, SQL injection, secrets, file/network/process use
- Performance regressions
Do NOT suggest SQL Server missing-index DMV recommendations — they are folklore in
this codebase and explicitly unwanted.
The PR branch is already checked out in the working directory.
Use `gh pr comment` for top-level feedback, and
`mcp__github_inline_comment__create_inline_comment` (with confirmed: true) for
specific line-level issues. Only post GitHub comments — do not emit review text as
plain messages.
claude_args: |
--allowedTools "mcp__github_inline_comment__create_inline_comment,Bash(gh pr comment:*),Bash(gh pr diff:*),Bash(gh pr view:*)"