What problem does it solve? Manually implementing a backlog of GitHub issues one at a time is slow and repetitive. This Skill automates the entire cycle: it fetches eligible issues, spawns parallel worker agents to implement each one, creates draft PRs, runs self-reviews, and reports consolidated results. ## Core Features & Use Cases - Issue Filtering & Validation: Fetches issues via the GitHub CLI, excludes items with existing PRs or AI labels, and skips underspecified issues missing acceptance criteria. - Parallel Agent Orchestration: Creates a team of worker agents (configurable concurrency up to 5) that each implement an issue on a conventional branch, lint, commit, and open a draft PR. - Self-Review & Reporting: Each worker reviews its own diff for bugs, security issues, and scope creep, then reports PR URLs and concerns back for a final summary table. - Use Case: You have five groomed backlog issues for a WordPress project. Invoke the Skill with the issue numbers, confirm the list, and receive five draft PRs with self-review notes and a results summary. ## Quick Start Ask the agent to run sprint-worker on issues 8601, 8590, and 8575 to implement them in parallel and open draft PRs.
