Skip to content

Works much worse since gpt 5.4 and codex performs remarkably better with the same model #2649

Works much worse since gpt 5.4 and codex performs remarkably better with the same model

Works much worse since gpt 5.4 and codex performs remarkably better with the same model #2649

Workflow file for this run

name: Close issue/PR on adding invalid label
# **What it does**: This action closes issues that are labeled as invalid in the repo.
on:
issues:
types: [labeled]
pull_request_target:
types: [labeled]
permissions:
contents: read
issues: write
pull-requests: write
jobs:
close-on-adding-invalid-label:
if:
github.repository == 'github/copilot-cli' && github.event.label.name ==
'invalid'
runs-on: ubuntu-latest
steps:
- name: Close issue
if: ${{ github.event_name == 'issues' }}
env:
GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
URL: ${{ github.event.issue.html_url }}
run: gh issue close $URL
- name: Close PR
if: ${{ github.event_name == 'pull_request_target' }}
env:
GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
URL: ${{ github.event.pull_request.html_url }}
run: gh pr close $URL