Development plugins

Align Rails load_defaults across versions with safe, incremental config changes

1 skill

Consistent orchestration patterns for Claude Code dynamic workflows, combining Anthropic's own docs with a GoF-style pattern catalog as a semantic vocabulary layer.

1 skill

Fork of Anthropic's skill-creator. Creates, tests, and iteratively improves Claude Code skills via eval loops, benchmarking, blind A/B agents, and description optimization.

1 skill

Session workflow guardrails, active on install: a session-start situation report, a stop gate that runs the tests before Claude finishes, type errors fed back after each edit, git footgun protection, a branch-first nudge, and a subagent audit log.

hooks

Forces a deliberate design decision before non-trivial code is written, a test-backed process before code is restructured, a structured review of someone else's abstractions, and an audit of what was actually built. Ships the patterns-and-refactoring skill: 22 GoF patterns, 66 refactoring techniques, 22 code smells, enterprise/architectural/DDD/distributed/concurrency/functional/frontend catalogues, anti-patterns, testing patterns, decision templates and framework idioms for 10 stacks.

1 skill

Unofficial skill that turns the claude.dev engineering blog into working guidance for Claude Code and the Claude API, refreshed from the site every hour

1 skill

Add local semantic search to an app with Chassis: vectors in one file, searched in-process, no server.

1 skill

Implement, upgrade, or debug AG Grid in JavaScript/TypeScript frameworks using its MCP server. For Java-authored JWebMP grids, use jwebmp-aggrid.

1 skill

Integrate external REST/GraphQL APIs or webhooks, including authentication, retries, rate limits, and request/response mapping.

1 skill

Convert Azure ARM templates to Terraform HCL configuration with full resource mapping and dependency preservation

1 skill

Model/framework benchmark, profile, paired confirmation and review with the inference-opt harness.

1 skill

Evidence-first engineering protocol for AI coding agents. A bundled tool runs the checks and names the result, so the model never grades its own work.

1 skill

The new-modeler skill: build a Miragon modeler end to end from a runnable template, on @miragon/modeler-api and @miragon/modeler-shell.

1 skill

The mushin usage guide for AI coding agents, as the /mushin:guide skill. Prints the guide that matches the installed mushin (0.16 or later).

1 skill

Continue another agent's chat (Claude Code, Codex, Cursor, OpenCode) and delegate work to them as background sub-agents under your model policy.

1 skill

Build native iOS and Android apps as verifiable twins on Duet: pure reducers, SwiftUI and Jetpack Compose shells, record/replay fixtures checked byte-for-byte in CI, and the duet CLI.

1 skill

GraphQL LSP plugin powered by graphql-analyzer

TypeScript native LSP plugin powered by tsgo

Побудова застосунків: Claude API, хірургічний рефакторинг великих файлів, E2E-тести, PWA/APK, локальні моделі в браузері. (5 скілів)

ggen language server: live diagnostics, completion, hover, definitions, references, document/workspace symbols, formatting, and repair-route code actions for ggen law surfaces (RDF .ttl/.nt/.nq, SPARQL .rq/.sparql, Tera .tera, and ggen.toml).

An automatic issue-to-merge workflow where an agent builds and people verify.

Expert agents for TRON DApp development — TronWeb usage patterns, Transatron fee optimization, and Privy embedded wallet integration

React/TypeScript + Go enforcement harness. Hooks enforce accessibility, React/TypeScript correctness, styling, env/data rules, warning-free tests and CI, security, PR feedback completeness, worktree safety, and lockfile parity. Skills cover planning, TDD, review, dependency upgrades, resilience, PR reviewability, visual planning, status summaries, CI-green shipping, and handoff. Includes Builder.io and Cursor Team Kit workflows. POSIX-friendly.

200 skills · 13 agents · hooks

Specialist agents, grumpy reviewers, and slash commands for project planning, scoping, implementation, and code review. Transforms Claude Code into a multi-agent development team.

66 commands · 33 agents · hooks

Full-stack developer toolkit for Claude Code with 48 commands, 2 agents, and 41 skills

44 skills · 46 commands · 2 agents · 1 MCP server

Pragmatic development toolkit — coding conventions, review agents, and workflow hooks that evolve through daily practice

35 skills · 13 agents

Set of setup tools, skills and agents for Spryker project development.

37 skills · 6 agents · hooks

Lens-based review skills and scaffolding skills for an agentic engine, backed by its spec

41 skills · 1 agent

Production-tested SKILL.md workflows for auditing, improving, and shipping repos with AI coding agents. Covers audits, security sweeps, publish gates, releases, memory handoffs, skill extraction, and decision-ready briefs

39 skills · hooks

Unified development workflow - design, planning, TDD, debugging, git protection, verification, review, and shipping

38 skills · 1 agent · hooks