By kenkenmain
Hook-driven subagent architecture with auto-chaining, gate enforcement, full context isolation, and no external plugin dependencies
Configure complexity scoring, models, and workflow options
Launch a multi-phase debugging workflow with parallel exploration, solution proposals, ranking, implementation, review, and documentation
Start a subagent workflow using Claude agents only (no Codex MCP)
Start a subagent workflow with Codex MCP defaults (use dispatch-claude for Claude-only)
Fast dispatch - streamlined 4-phase workflow with combined plan+implement+review (Claude-only, no Codex MCP)
Analyzes existing codebase patterns and conventions, provides architecture blueprint with component designs and data flows.
Analyzes explore findings and determines implementation approach with trade-offs. Use proactively after codebase exploration to evaluate strategies before planning.
Reviews session for learnings and updates CLAUDE.md with concise context for future sessions.
Use proactively to review plans, implementation, tests, and final output using Claude reasoning (no Codex MCP)
Reviews code for bugs, logic errors, style violations, and adherence to project conventions.
Load, merge, and validate subagents configuration from global and project files
Manage workflow state in .agents/tmp/state.json with atomic writes
Ralph-style orchestrator loop — dispatches each phase as a subagent, hooks enforce progression
Modifies files
Hook triggers on file write and edit operations
Uses power tools
Uses Bash, Write, or Edit tools
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
Claude Code plugin marketplace for development workflows.
# Add the marketplace
claude plugin marketplace add kenkenmain/ken-cc-plugins
# Install plugins
claude plugin install superpowers-iterate@ken-cc-plugins
| Plugin | Phases | Parallelism Style | Model Requirements | Best For | Example Tasks |
|---|---|---|---|---|---|
| kenken | 4 (Plan, Implement, Test, Final) | Sequential stages, no parallel agents | Claude only | Quick tasks, simple iteration | "Add OAuth2 login", "Set up GitFlow repo" |
| superpowers-iterate | 9 (loop phases 1-8 until clean) | Sequential with iterative refinement | Codex MCP (full) or Claude-only (lite) | Quality-obsessed iteration, convergence to zero issues | "Refactor auth module with TDD", "Add comprehensive error handling" |
| minions | 5 (launch) / 15 (superlaunch) / 5 (cursor) / 2 (review) | Parallel explorers + parallel reviewers (6 specialists) | Claude only (no Codex MCP) | Thorough coverage, personality-driven review teams | "Build payment pipeline (superlaunch)", "Review-fix existing code (review)" |
| subagents | 5-15 (profile-based) | Wave-based parallel agents, complexity-routed tasks | Codex MCP (default) or Claude-only | Complex multi-file features, configurable pipelines | "Add full API with tests (standard)", "Quick config fix (minimal)" |
| ants | 6 (A0-A5), persistent loop variant | Self-organizing task pool + adversarial review team (5 sentinels) | Claude only, Agent Teams | Adversarial quality, fast iteration, concurrent builds | "Add caching layer (swarm)", "Fix all lint warnings (pswarm)" |
Quick decision guide:
Hook-driven subagent architecture for complex task execution with parallel agents and file-based state.
claude plugin install subagents@ken-cc-plugins
# Standard workflow (13 phases)
/subagents:dispatch <task> # Codex MCP defaults
/subagents:dispatch-claude <task> # Claude-only mode
/subagents:init <task> # With git worktree isolation
# Fast workflow (4 phases)
/subagents:fdispatch <task> # Fast dispatch (Codex MCP)
/subagents:fdispatch-claude <task> # Fast dispatch (Claude-only)
# Debug workflow (6 phases)
/subagents:debug <task> # Multi-phase debugging
# Management
/subagents:status # Check progress
/subagents:stop # Stop workflow gracefully
/subagents:resume # Resume from checkpoint
/subagents:teardown # Commit, push, create PR
/subagents:configure # Configure settings
Pipelines: Standard (Explore -> Plan -> Implement -> Test -> Final) | Fast (Plan -> Implement -> Review -> Complete) | Debug (Explore -> Propose -> Aggregate -> Implement -> Review -> Document)
Features: Complexity-routed task agents, hybrid test-alongside-code, Codex/Claude dispatch modes, pipeline profiles (minimal/standard/thorough), fast dispatch (4-phase variant), multi-phase debug workflow
See plugins/subagents/README.md for full documentation.
9-phase iterative development workflow. Phases 1-8 loop until Phase 8 finds zero issues.
claude plugin install superpowers-iterate@ken-cc-plugins
/superpowers-iterate:iterate <task> # Full mode (requires Codex MCP)
/superpowers-iterate:iterate --lite <task> # Lite mode (no Codex required)
/superpowers-iterate:iterate-status # Check progress
Phases: Brainstorm -> Plan -> Plan Review -> Implement -> Review -> Test -> Simplify -> Final Review -> Codex Final
Prerequisites: superpowers plugin, code-simplifier plugin, Codex MCP (full mode only)
See AGENTS.md for workflow architecture, modes, and state management details.
Personality-driven multi-pipeline development workflow with loop-back issue resolution. Claude-only (no Codex MCP required).
claude plugin install minions@ken-cc-plugins
/minions:launch <task> # Standard 5-phase workflow
/minions:superlaunch <task> # Claude-only 15-phase thorough pipeline
/minions:cursor <task> # Cursor-inspired per-task commit workflow
/minions:review <task> # Review-fix iteration pipeline
Launch Pipeline: Explore (4x parallel haiku) -> F1 Scout -> F2 Build -> F3 Review (6 parallel reviewers) -> F4 Ship
npx claudepluginhub kenkenmain/ken-cc-plugins --plugin subagents4-stage iterative development workflow (PLAN → IMPLEMENT → TEST → FINAL)
Orchestrates iterative 9-phase workflow with configurable models and full/lite modes.
Personality-driven 4-phase workflow (launch), Claude-only 15-phase thorough pipeline (superlaunch), Cursor-inspired pipeline (cursor), and review-fix iteration pipeline (review)
Ant-colony themed swarm workflow with Agent Teams delegate mode featuring team creation with shared task lists, TeammateIdle task assignment, TaskCompleted quality gates and state advancement, dual-channel communication (file-based artifacts as source of truth plus SendMessage live coordination overlay for 16 agents), social swarm (sswarm) with competing parallel agents coordinated via task dependency chains, persistent pswarm mode with signal-flag-driven task graph recreation, dual-track build, adversarial review (6 specialist sentinels: correctness, security, perf, style, testing, docs), inline A4 verdict evaluation, post-build simplifier, A0 explore-aggregator, debug pipeline for bug investigation, self-improvement pipeline for iterative review-fix, graceful shutdown, plan approval, lint-on-save, HTTP webhooks, opt-in WebSearch support, and structured completion summaries for all five pipelines
This skill should be used when the model's ROLE_TYPE is orchestrator and needs to delegate tasks to specialist sub-agents. Provides scientific delegation framework ensuring world-building context (WHERE, WHAT, WHY) while preserving agent autonomy in implementation decisions (HOW). Use when planning task delegation, structuring sub-agent prompts, or coordinating multi-agent workflows.
Claude Code の SubAgent(agents/*.md)を正しく定義するためのベストプラクティスガイド。 YAML frontmatter、ツール選択、3-Phase 構造、コンテキスト受け渡し、アンチパターンを網羅。 Use when: agents/*.md を書く、SubAgent 定義を改善する、エージェントの動作が想定外、 コンテキストが渡らない、ツール選択に迷う。 Triggers: "subagent", "agent definition", "agents/*.md", "エージェント定義", "サブエージェント", "3-Phase", "context passing", "コンテキスト渡し", "tool selection", "ツール選択", "subagent_type", "bypassPermissions"
Multi-agent collaboration plugin for Claude Code. Spawn N parallel subagents that compete on code optimization, content drafts, research approaches, or any problem that benefits from diverse solutions. Evaluate by metric or LLM judge, merge the winner. 7 slash commands, agent templates, git DAG orchestration, message board coordination.
Multi-agent orchestration framework for Claude Code. Routes tasks to specialized Haiku/Sonnet subagents while Opus orchestrates — inspired by speculative decoding. Includes 10 specialized heads, environment preflight checks, and ~50% API cost reduction.
Sub-agent runner — runs agent definitions on Codex, Claude Code, Cursor CLI, or Gemini CLI from any host AI tool.
Parallel task orchestration for AI coding agents - dispatch work to Codex or Claude Code workers in isolated git workspaces