By dynamic-dome
Multi-model agent team orchestration. Routes tasks to the best model (Claude, Codex, etc.) with isolated git worktrees. Supports single-agent, dual-mode (plan-implement-review), and custom pipelines.
Autonomous code implementation agent. Reads a PLAN.md and implements every step: writes code, creates files, runs tests. Designed for handoff from a planner agent. Uses the most cost-efficient model available.
Senior architect agent that analyzes tasks and creates detailed step-by-step implementation plans. Used as the planning phase in multi-model pipelines before handing off to an implementer agent.
Code review agent that evaluates implementation changes against a plan. Checks correctness, security, quality, and test coverage. Outputs a structured PASS/FAIL verdict. Used as the final quality gate in multi-model pipelines.
Task routing agent that decides which model or pipeline mode to use for a given task. Routes to single-agent (Claude or Codex), dual-mode (plan-implement-review), or custom agent teams based on task complexity and type.
Spawnt N parallele Codex-Agents via tmux, orchestriert von Claude Opus 4.6. Jeder Agent laeuft in eigener Worktree mit konfigurierbarem Modell und Prompt.
Route implementation tasks to OpenAI Codex in a sandboxed container for autonomous code churn
OPTIONAL Docker-Sandbox-Modus fuer multi-model-orchestrator. Standardmodus des Plugins ist tmux+Worktrees (siehe scripts/orchestrator.sh) — dieser Skill ist eine alternative Isolations-Schicht fuer destruktive oder besonders sensible Tasks. Triggert nur bei explizitem Wunsch nach Container-Isolation.
REDIRECT: Use agentic-os:research-pipeline instead. This is an alias to avoid duplication.
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.
inception-sandbox/)Identitaet (Stand 2026-04-30): Plugin-Name ist
multi-model-orchestrator. Der Repo-Pfad heisst aus historischen Gruendeninception-sandbox— ein Rename wuerde Cache, Marketplace-Eintrag, lokale Pfade und Submodule-Referenzen brechen ohne Public-API-Mehrwert. Andere Plugins rufen diesen uebermulti-model-orchestrator:codex-swarmu. a. auf, der Manifest-Name ist die einzige externe API. Volle Erklaerung:CLAUDE.mdSektion "Identitaet".
Multi-Model-Orchestrierung: Claude Code + OpenAI Codex via tmux + Git Worktrees.
Optionaler Docker-Sandbox-Modus via skills/inception/ fuer Faelle, in denen Worktree-Isolation nicht reicht.
Claude und Codex laufen lokal in separaten tmux-Sessions. Jeder Agent arbeitet in einer eigenen Git Worktree fuer Isolation. Keine API-Keys noetig — OAuth reicht.
┌─────────────────────────────────────────────────┐
│ tmux session "inception" │
│ │
│ Window "claude" Window "codex" │
│ ┌──────────────┐ ┌──────────────┐ │
│ │ claude -p │ │ codex │ │
│ │ --dangerously │ │ --full-auto │ │
│ │ -skip-perms │ │ --quiet │ │
│ │ │ │ │ │
│ │ Worktree A │ │ Worktree B │ │
│ └──────────────┘ └──────────────┘ │
│ │ │ │
│ └───── shared via ──────┘ │
│ file copy │
│ (PLAN.md, diffs) │
└─────────────────────────────────────────────────┘
inception-sandbox/
├── CLAUDE.md # Projekt-Instruktionen
├── README.md # Diese Datei
├── scripts/
│ ├── orchestrator.sh # Multi-Model Orchestrator (single/dual mode)
│ ├── send-prompt.sh # Prompt an Agent senden via tmux
│ └── read-output.sh # Output aus Agent lesen via tmux
├── skills/
│ ├── inception/SKILL.md
│ ├── codex-worker/SKILL.md
│ └── research-pipeline/SKILL.md
├── research/ # Perplexity/NotebookLM Recherche-Ergebnisse
└── output/ # Ergebnisse der Orchestrator-Laeufe
# Voraussetzungen: tmux, claude CLI (Max Plan), codex CLI (Desktop App)
# Einfach: Claude loest Task alleine
./scripts/orchestrator.sh --prompt "Fix the login bug"
# Codex fuer Bulk-Arbeit
./scripts/orchestrator.sh --agent codex --prompt "Refactore alle Tests zu pytest"
# Dual-Mode: Claude plant → Codex baut → Claude reviewed
./scripts/orchestrator.sh --mode dual --prompt "Add pagination to the API"
# Auf beliebigem Repo arbeiten
./scripts/orchestrator.sh --mode dual --repo ~/projects/myapp --prompt "Add auth"
| Modus | Ablauf | Wann nutzen |
|---|---|---|
single --agent claude | Claude alleine | Review, Planung, Security |
single --agent codex | Codex alleine | Refactoring, Linting, Bulk |
dual | Claude → Codex → Claude | Komplexe Features |
PLAN.md mit Architektur + Schrittenoutput/sudo apt install tmux (WSL2) oder via Git BashClaude Max Plan und Codex Desktop App authentifizieren via OAuth/Browser, nicht via API-Key. Docker-Container koennten sich nicht einloggen. tmux + Git Worktrees bieten ausreichende Isolation fuer lokale Entwicklung.
npx claudepluginhub dynamic-dome/inception-sandbox --plugin multi-model-orchestratorAdapter-agnostischer 4-Rollen-Loop (Orchestrator/Researcher/Builder/Judge) mit persistierenden State-Dateien pro Projekt.
Autonomous improvement loop for Claude Code projects — decomposes goals into tasks, executes them via Worker/Judge subagents, mines session logs for error patterns, and generates evidence-based CLAUDE.md improvement suggestions.
Interactive Claude Code workshop plugin — two modes: /workshop guide (moderator co-pilot) and /workshop learn (self-paced tutor). 17 modules (5+5+7) across 3 blocks covering foundations, ecosystem, and advanced topics.
Autonomer Meta-Agent-Orchestrator: Ein Opus 4.6 Agent der Haiku-Brainstormer, NotebookLM-RAG und Codex-Instanzen orchestriert um komplexe Aufgaben autonom zu loesen.
Suite of 7 skills, 7 slash commands, and 5 read-only checker hooks that turns rough ideas into agentic workflow blueprints, runs adversarial review-fix-verify loops, audits ecosystems as dynamic graphs, curates raw inbox material into wiki pages, runs MCP/DCO actions safely, and gates skill promotion. Built from the wiki/patterns/agentic-workflow-blueprints reference.
Comprehensive UI/UX design plugin for mobile (iOS, Android, React Native) and web applications with design systems, accessibility, and modern patterns
Multi-model consensus engine integrating OpenAI Codex CLI, Gemini CLI, and Claude CLI for collaborative code review and problem-solving.
Ultra-compressed communication mode. Cuts ~75% of tokens while keeping full technical accuracy by speaking like a caveman.
Standalone image generation plugin using Nano Banana MCP server. Generates and edits images, icons, diagrams, patterns, and visual assets via Gemini image models. No Gemini CLI dependency required.
Write feature specs, plan roadmaps, and synthesize user research faster. Keep stakeholders updated and stay ahead of the competitive landscape.
Curate auto-memory, promote learnings to CLAUDE.md and rules, extract proven patterns into reusable skills.