A comprehensive agile development and repository management skill for AI coding assistants. Install it once and get sprint planning, issue resolution, release management, code review, CI/CD templates, and much more — all driven by git and gh CLI commands.
npx skills add yabasha/agile-workflowOr manually copy SKILL.md into your project's .skills/ directory or AI assistant's skill/instruction folder.
End-to-end workflow for batch-fixing GitHub issues in parallel — discover open issues, prioritize, spawn isolated developers on worktrees, create PRs, resolve review comments, merge in dependency order, and clean up.
- Sprint Planning — Create milestones, assign issues, capacity planning heuristics
- Backlog Grooming — Issue templates (bug, feature, chore, spike), label taxonomy, grooming process
- Story Point Estimation — Fibonacci scale with a programmatic estimation algorithm
- Issue Triage — Decision tree for categorizing and prioritizing incoming issues with auto-labeling
- Sprint Retrospectives — Automated metrics collection, retro report template, action item tracking
- Kanban Board Management — GitHub Projects setup, WIP limits, lead/cycle time metrics
- Epic & Milestone Tracking — Epic decomposition, milestone progress monitoring
- Status Reports & Standups — Daily standup and sprint progress report generators
- Definition of Done — Universal and per-type checklists with automated verification
- Branch Strategy — GitFlow, GitHub Flow, and Trunk-Based Development with a decision tree
- Conventional Commits — All 11 types, scopes, breaking changes, commitlint enforcement
- Release Management — SemVer decision tree, tagging, changelog generation, GitHub Releases, rollback procedures
- Repository Setup —
.gitattributes,.gitignore, CODEOWNERS, branch protection rules, issue/PR templates - Repository Health — Stale branch detection, orphaned PRs, large file detection, security alerts, repo statistics
- Monorepo Management — Workspace configs (bun/pnpm/npm/yarn), changesets, selective CI
- Git Hooks — Husky + lint-staged + commitlint + pre-push tests + post-merge auto-install
- Advanced Git Operations — Rebase, cherry-pick, bisect, worktree, stash, reflog, submodules vs subtrees
- GitHub Actions workflow templates (CI, PR checks, release, dependency updates)
- Test gates and coverage thresholds
- Security scanning (Dependabot, gitleaks, secret scanning)
- Deployment strategy decision tree
- Review checklist (correctness, tests, security, performance, style, dependencies)
- Comment templates (suggestion, required change, nitpick, question)
- Automated first-pass review workflow
- SVN-to-Git Migration — Author mapping, clone with history, post-migration cleanup, verification
- Git Error Recovery Cheat Sheet — Quick-reference table for common recovery operations
This skill works with any AI coding assistant that has shell access:
| Platform | Support |
|---|---|
| Claude Code | Full support (parallel agents, worktrees, team orchestration) |
| Cursor | Full support |
| Windsurf | Full support |
| GitHub Copilot | Full support |
| Codex | Full support |
| OpenClaw | Full support |
| Aider | Full support |
| Any LLM + terminal | Full support |
All commands use standard git and gh (GitHub CLI) — no proprietary APIs or platform-specific dependencies.
- Git 2.20+
- GitHub CLI (
gh) 2.0+ - A GitHub repository
Once installed, ask your AI assistant to perform any of the supported workflows:
> fix all open github issues
> plan the next sprint
> estimate story points for ungroomed issues
> set up conventional commits with husky
> create a release for v2.0.0
> review PR #42
> clean up stale branches
> triage new issues
> generate a sprint retrospective report
The skill includes a module dispatcher that routes your intent to the correct workflow automatically.
agile-workflow/
SKILL.md # The skill (all instructions, templates, and commands)
README.md # This file
MIT
Bashar Ayyash
- Portfolio: yabasha.dev
- GitHub: @yabasha