name: github-project-week-issues description: Use when creating or planning weekly GitHub Issues from github-projects/week*_issues_complete.csv or similar weekly CSV files and adding them to a GitHub Project. Trigger for tasks that mention GitHub Projects, weekly issues, week_n labels, CSV issue files, sprint/week setup, repository/project issue import, or adding the corresponding week's issues to a project with gh CLI.

GitHub Project Week Issues

Use this skill to turn a weekly CSV file into a reviewed GitHub issue import for a repository and GitHub Project.

Workflow

1. Identify the target week and source file.
   • Prefer an explicit --week or --csv from the user.
   • If the user says "해당 주차", inspect github-projects/ and choose the matching week*_issues_complete.csv.
   • Use scripts/plan_week_issues.py to parse the CSV and infer labels.
2. Inspect live GitHub state before changing anything:
   • target repository issue list
   • target repository labels
   • target GitHub Project fields, options, and current items
   • repository folder/file structure when issue titles reference paths, tests, or modules
3. Produce a decision-complete plan first unless the user explicitly asked to apply immediately.
   • Include label additions/reuse, title/body rules, duplicate prevention, project status, and verification commands.
   • Do not create issues while the user is still asking for planning or review.
4. Apply only after approval or an explicit "추가해줘/실행해줘" style request.
   • Create missing labels.
   • Create non-duplicate issues.
   • Add created or reused issues to the project.
   • Set project status/fields when field IDs and option IDs are known.
5. Verify the result.
   • Confirm issue count, labels, project membership, and status field values.
   • Report skipped duplicates separately from created issues.

Helper Script

Run the helper from the target repository root or pass explicit paths:

python3 .codex/skills/github-project-week-issues/scripts/plan_week_issues.py --week 6 --repo OWNER/REPO --project-url https://github.com/users/OWNER/projects/1

Useful modes:

# Markdown dry-run plan from github-projects/week6_issues_complete.csv
python3 .codex/skills/github-project-week-issues/scripts/plan_week_issues.py --week 6 --repo OWNER/REPO

# JSON plan for custom automation
python3 .codex/skills/github-project-week-issues/scripts/plan_week_issues.py --csv github-projects/week11-12_issues_complete.csv --format json

# Apply labels/issues and add them to a project after approval
python3 .codex/skills/github-project-week-issues/scripts/plan_week_issues.py --week 6 --repo OWNER/REPO --project-url https://github.com/users/OWNER/projects/1 --status Todo --apply

The script defaults to dry-run. --apply requires --repo; project addition requires --project-owner and --project-number or --project-url.

GitHub Inspection

Use gh first:

gh repo view OWNER/REPO --json nameWithOwner,description,defaultBranchRef
gh issue list --repo OWNER/REPO --state all --limit 1000 --json number,title,state,labels,url
gh label list --repo OWNER/REPO --limit 500 --json name,description,color
gh project view PROJECT_NUMBER --owner OWNER --format json
gh project field-list PROJECT_NUMBER --owner OWNER --limit 100 --format json

For project items, use GraphQL when gh project item-list is not enough:

gh api graphql -f query='
query($owner:String!, $number:Int!) {
  user(login:$owner) {
    projectV2(number:$number) {
      id
      items(first:100) {
        nodes {
          id
          content { ... on Issue { title number url repository { nameWithOwner } } }
          fieldValues(first:20) { nodes { ... on ProjectV2ItemFieldSingleSelectValue { name field { ... on ProjectV2SingleSelectField { name } } } } }
        }
      }
    }
  }
}' -f owner=OWNER -F number=1

Use organization(login:$owner) instead of user(login:$owner) for organization-owned projects.

Issue Rules

• Preserve CSV titles unless the existing project uses a clear week prefix convention.
• Treat the Korean prefix before - as a category hint:
  • 공통 -> category_common
  • 학습 -> category_learning
  • 구현 -> category_implementation
  • 테스트 -> category_test
  • 문서 -> category_docs
  • 운영 -> category_ops
  • 버그 -> category_bug
  • 기능 -> category_feature
• Always add the week label, using week_6 or week_11_12 for ranges.
• Keep label count restrained. Prefer existing labels with matching meaning.
• Use exact normalized title matching for duplicate prevention first; then inspect close matches manually when the project already has weekly imports.
• For empty CSV content, create a short body that records the source file, week, category, and original CSV title instead of inventing details.

References

• Read references/gh-project-notes.md when project field/status editing is needed.