By gijungkim
Automatically resolve Dependabot PRs - assesses risk, tests locally, merges safe ones, fixes breaking changes, and closes incompatible ones with clear explanations.
npx claudepluginhub gijungkim/resolve-prs --plugin resolve-prsA Claude Code skill that automatically resolves Dependabot PRs on your GitHub repos.
Instead of manually reviewing each Dependabot PR, this skill:
/plugin marketplace add GijungKim/resolve-prs
/plugin install resolve-prs@resolve-prs
Or manually copy skills/resolve-prs/SKILL.md to ~/.claude/skills/resolve-prs/SKILL.md.
# Resolve PRs on the current repo
/resolve-prs
# Target a specific repo
/resolve-prs owner/repo
# Process all git repos in the current directory (parallel, one agent per repo)
/resolve-prs --all
# Assess PRs without taking action (just report)
/resolve-prs --dry-run
# Combine flags
/resolve-prs --all --dry-run
gh CLI authenticated with repo accessThe skill includes a growing knowledge base of common breaking changes it can detect and handle:
| Pattern | What it does |
|---|---|
| ESLint 8 -> 9+ | Detects flat config requirement, closes if not migrated |
| Jest major bumps | Checks framework preset compatibility (jest-expo, etc.) |
| react / react-dom mismatch | Closes if versions would diverge |
| react-native-mmkv 3 -> 4 | Migrates new MMKV() to createMMKV(), fixes .delete() -> .remove(), adds jest mock |
| Expo SDK pins | Respects Expo version constraints |
| Peer dependency conflicts | Detects and closes incompatible bumps |
When the skill fixes a novel breaking change not in its knowledge base, it automatically records the pattern. This means it gets smarter the more you use it. Patterns are kept generalizable (not project-specific), capped at 30 per category, and consolidated when they get stale.
--all worksWhen you pass --all, the skill:
This is useful when you have a directory of projects (e.g., ~/projects/) and want to batch-resolve Dependabot PRs across all of them.
--dry-run worksDry-run mode goes through the full assessment workflow (listing PRs, fetching changelogs, categorizing risk, testing locally) but skips all merge/close/fix actions. The report uses "Would merge", "Would close", "Would fix & merge" so you can review what it plans to do before running it for real.
MIT
Battle-tested Claude Code plugin for engineering teams — 48 agents, 184 skills, 79 legacy command shims, production-ready hooks, and selective install workflows evolved through continuous real-world use
Automates browser interactions for web testing, form filling, screenshots, and data extraction
Tools to maintain and improve CLAUDE.md files - audit quality, capture session learnings, and keep project memory current.
Comprehensive skill pack with 66 specialized skills for full-stack developers: 12 language experts (Python, TypeScript, Go, Rust, C++, Swift, Kotlin, C#, PHP, Java, SQL, JavaScript), 10 backend frameworks, 6 frontend/mobile, plus infrastructure, DevOps, security, and testing. Features progressive disclosure architecture for 50% faster loading.
Complete collection of battle-tested Claude Code configs from an Anthropic hackathon winner - agents, skills, hooks, rules, and legacy command shims evolved over 10+ months of intensive daily use
Comprehensive .NET development skills for modern C#, ASP.NET, MAUI, Blazor, Aspire, EF Core, Native AOT, testing, security, performance optimization, CI/CD, and cloud-native applications