菜单
GitHub integration via gh CLI. List issues, create PRs, comment on issues. Use for agents that bridge GitHub and the Workshop board.
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
| name | github |
| description | GitHub integration via gh CLI. List issues, create PRs, comment on issues. Use for agents that bridge GitHub and the Workshop board. |
| license | MIT |
| metadata | {"author":"joshrotenberg","version":"1.0"} |
Bridge GitHub issues and PRs with the Workshop board using the gh CLI.
Verify authentication before any operation:
gh auth status
# List open issues with a label
gh issue list --label "agent" --state open --json number,title,body,labels --limit 50
# Read a single issue
gh issue view 42 --json number,title,body,labels,comments
Always use --json for structured output.
When syncing issues to the board, use the ID convention gh_<number> for dedup:
# Check if already on the board before adding
board()
# Add a new work item from an issue
add_work(id: "gh_42", title: "Issue title", type: "triage", priority: 3,
spec: "Issue body text here")
Skip issues that already have a board item (any status).
After work is completed on a branch:
# Create PR linking back to the issue
gh pr create --title "fix: description" --body "Closes #42" --head branch-name --base main
Use Closes #N in the PR body to auto-close the issue on merge.
# Status update on an issue
gh issue comment 42 --body "Picked up by agent, work item gh_42 created."
# Label management
gh issue edit 42 --add-label "in-progress"
gh issue edit 42 --remove-label "agent"
Declarative multi-stage workflows. Define staged pipelines where each stage feeds its result to the next. Use for structured multi-agent tasks.
Autonomous board worker. Receives work items, implements them, reports results. Designed for agents that poll the board and execute tasks independently.
Work board driven development. Post typed work items with dependencies, agents poll by role. Use for multi-step features where ordering matters.
Multiple LLM backends in one workshop. Claude and Codex agents working together. Use when different models have different strengths.
Scheduled agent for periodic checks. Use for CI monitoring, PR scanning, health checks, or any polling-based workflow.
Coordinate a team of AI agents to build software. Break tasks down, create specialist agents from profiles, delegate via ask/cast, review results, and clean up. Use when you need to orchestrate multi-agent workflows.