Development plugins

iOS and tvOS development with SwiftUI, MVVM, Clean Architecture, 14 specialized agents, 27 comprehensive skills, automated quality gates, hooks system, one-command setup (/ios-init), beads integration, offline sync, push notifications, tvOS focus navigation, accessibility testing, performance profiling, and dual memory systems

27 skills · 6 commands · 14 agents · hooks

Token-optimized software development plugin with 22 skills, 16 agents, and governance gates for full lifecycle development from requirements to shipping

22 skills · 16 agents · hooks

AST-aware code transformer for AI agents. 140+ refactoring operations across 100+ languages with undo/redo, preview mode, and intelligent guardrails.

26 skills · 3 commands · 3 agents · 1 MCP server

Flutter development with GetX state management, Clean Architecture, multi-agent orchestration, quality gates, comprehensive testing patterns, navigation, i18n, performance optimization, and accessibility patterns

16 skills · 5 commands · 9 agents

One-path risk-budgeted workflow with selective agents, single-pass verification, and explicit completion gates.

9 skills · 17 agents · 1 MCP server · hooks

Async-offload companion. Delegate reviews, audits, validations, research, summarization, and bulk RAG ingest to vornik-managed agents over MCP; results re-enter Claude's context via a SessionStart digest hook. Per-repo scoping (migration 75) so one project's RAG cleanly serves multiple repos. Also bundles operator lifecycle skills — configure-vornik, validate-install, troubleshoot-vornik, and report-problem — so your assistant drives Vornik's own tooling (config validate/reload, the published reference architecture, the doctor, the failure-class playbook, and the anonymized problem reporter) instead of improvising.

6 skills · 19 commands · 1 MCP server · hooks

Comprehensive Clerk authentication plugin with OAuth (19 providers), organizations, billing, MFA, Next.js integration, and Supabase sync for building secure multi-tenant applications

10 skills · 2 commands · 15 agents

Lightweight development workflow with plan/run/fresh pattern, git workflow integration (branch-per-plan, PR creation, PR feedback), on-demand skills, and specialized agents. Optimized for efficiency with minimal hooks and direct execution.

19 skills · 6 agents · hooks

Machine learning training and inference pipeline using cloud GPUs (Modal, Lambda Labs, RunPod) with HuggingFace ecosystem - no local GPU required

8 skills · 2 commands · 16 agents

Comprehensive Supabase integration for AI applications with database, auth, storage, realtime, and vector search capabilities

8 skills · 2 commands · 15 agents

Queryable code intelligence - explore codebase structure without reading files

17 skills · 2 agents · 1 MCP server · hooks

Feature-centric development framework for Claude Code — brainstorm, plan, build, and verify with structured workflows

14 skills · 5 agents · hooks

Your organization's RepoQL uplinks from one remote connection - no local install

13 skills · 2 agents · 1 MCP server

Eval-first plugin builder: primitive decision records, generated eval suites with arming gates, sha256-frozen contracts, goal-loop builds in isolated worktrees, verified ships. The suite is the contract; green is the definition of done.

11 skills · 2 agents · hooks

Cross-model audit gate — 3-adapter (Claude Code + Gemini CLI + Codex), parliament protocol, 5 enforcement gates, 26 MCP tools, 3030 tests.

11 skills · 1 MCP server · hooks

Prompt engineering for current Claude models, Python and C++ reinforcement learning, Keras 3 deep learning, x402 payments, and OpenClaw setup: 6 skills, 4 commands and 2 agents.

6 skills · 4 commands · 2 agents

Vornik async-offload companion for Codex. Delegate reviews, audits, validations, research, summarization, and project RAG memory work to a vornik daemon over MCP. 0.12.0: mirrors the claude-code 0.14.0 operator lifecycle triad — `configure-vornik` and `troubleshoot-vornik` join `report-problem`, carrying the same config-resolution hazards (fallback chain, silently skipped VORNIK_CONFIGS_DIR, cwd trap, restart-vs-reload) and the same symptom-routed diagnostic ladder. Copies are identical to the claude-code originals except that they address Codex. 0.13.0: mirrors the claude-code 0.15.0 `whoami` MCP tool — since Codex has no SessionStart hook announcing the detected repo scope (unlike Claude Code), the delegate skill now points Codex at `mcp__vornik__whoami` to check project_id / default_repo_scope / effective_repo_scope on demand, useful when the operator runs several Codex sessions a day against different projects/repos. 0.13.1: README's MCP tool list gained `whoami`. 0.13.0 wired the delegate skill's guidance but left the tool out of the advertised list — on the client that needs it most, since Codex has no SessionStart hook announcing the detected scope, which is the entire reason whoami exists. 0.14.0: mirrors the claude-code 0.16.0 `report_problem` MCP verb — the remote-daemon report path. Codex users hit this harder than most, since a Codex session is often not on the daemon host at all. Collects nothing host-side; nothing is submitted without the user. 0.15.0: mirrors the claude-code 0.17.0 `validate-install` skill — checking an install against the published reference architecture, read-only, on resolved state rather than configured state. Copies are identical to the claude-code originals except that they address Codex. Also mirrors `vornik-docs` (where the documentation lives, CLI --help before recall, and the site map). 0.15.1: mirrors the claude-code 0.17.1 VORNIK_API_URL trap in configure-vornik — the CLI's target daemon defaults to localhost:8080, i.e. pr

8 skills · 1 MCP server

Core SDD workflow - spec-driven development from planning to implementation. Includes deep-spec, spec-to-task, feature-tdd, sdd-developer, feature-e2e-verifier, bug-fix, enhancement, and load-project-features skills. The enhancement workflow guides a thorough interview, an approved plan, implementation, and suitable verification.

8 skills · 1 MCP server

Multi-agent development workflow with five modes: plan, dev, full, auto, and bugfix. Uses Agent Teams for collaborative roles (Architect, PM, Developer, Tester, Reviewer) and a Researcher subagent for documentation lookups via Context7.

1 skill · 5 commands · 1 agent · hooks

Package-specific skills for integrating, tuning, diagnosing, optimizing, and migrating the scroll_spy Flutter package.

7 skills

Structured thinking, evidence-first planning, diagnosis, technical research, and documentation capture for AI agents.

6 skills · hooks

Ready GitLab MR review with discussions first, CI-first evidence, bounded agents, and one clean-pass approval.

1 skill · 4 agents · 1 MCP server

Persistent memory for Claude Code agents. Stores decisions, patterns, and knowledge in a structured graph database across sessions with auto-injection and enforcement.

5 skills · hooks

Generate amazing diagrams using 4 text-based formats (SVG, Mermaid, Excalidraw, D2) with intelligent format routing, 20+ diagram types, a shared design system with color theory, WCAG accessibility, and composition rules.

6 skills

Review product UI code against the Koobiq design system, migrate off deprecated APIs, upgrade the library and lint Koobiq usage.

5 skills · 1 agent

Behavior testing workflows for React and Angular contributors.

5 skills

Save, search, and switch AI coding context across 11 tools. Never lose context within or across AI tools.

4 skills · 1 MCP server

Build, playtest, and prove Godot games through MCP.

4 skills · 1 MCP server

Multi-agent pipeline for Manim Community Edition video generation. Transforms ideas, research papers, and math topics into animated videos via /manim-video.

1 skill · 1 command · 3 agents

Plugin development tools: scaffold new plugins, validate SKILL.md frontmatter, audit hooks for silent failures

2 skills · 2 commands