By ferni-ai
Iterative parallel orchestration combining Ralph Loop iteration with swarm agent parallelism. Spawns waves of agents, synthesizes results, and iterates until completion.
Cancel active Wave Orchestration
Clean up Wave Orchestration outputs
Show Wave Orchestration help and usage
Check if environment is ready for Wave Orchestration
Check current Wave Orchestration status
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.
Iterative parallel orchestration for Claude Code
Combines Ralph Loop (persistent iteration) with swarm parallelism (multiple agents) to create waves of parallel work that compound until completion.
TIME -->
Wave 1 Wave 2 Wave 3
[A] [B] [D] [E] [G]
[C] [F] DONE!
Each wave spawns parallel agents.
Each wave builds on prior results.
Iterate until complete.
# Step 1: Add the marketplace
/plugin marketplace add sethdford/wave-orchestration
# Step 2: Install the plugin
/plugin install wave-orchestration@wave-orchestration
git clone https://github.com/sethdford/wave-orchestration.git ~/wave-orchestration
/plugin install ~/wave-orchestration
# Check environment is ready
/wave-orchestration:wave-preflight
# Start wave orchestration with a goal
/wave-orchestration:wave "Build a REST API with user auth, CRUD endpoints, and tests" --max-iterations 15
# Check status
/wave-orchestration:wave-status
# Cancel if needed
/wave-orchestration:cancel-wave
| Command | Description |
|---|---|
/wave-orchestration:wave "goal" | Start wave orchestration |
/wave-orchestration:wave-status | Check current iteration and progress |
/wave-orchestration:cancel-wave | Stop the orchestration loop |
/wave-orchestration:cancel-wave --clean | Stop and clean up output files |
/wave-orchestration:wave-preflight | Verify environment is ready |
/wave-orchestration:wave-clean | Remove wave output files |
/wave-orchestration:wave-help | Show help and usage |
/wave-orchestration:wave "goal" [--max-iterations N] [--completion-promise TEXT] [--clean]
--max-iterations N — Stop after N iterations (default: 20)--completion-promise TEXT — Phrase that signals completion (default: DONE)--clean — Clear previous wave outputs before starting┌─────────────────────────────────────────────────────────────────┐
│ /wave-orchestration:wave "Build a REST API" │
└─────────────────────────────────────────────────────────────────┘
│
▼
┌─────────────────────────────────────────────────────────────────┐
│ setup-wave.sh │
│ ├── Creates .claude/wave-state.local.md (iteration: 1) │
│ └── Creates .claude/wave-outputs/ directory │
└─────────────────────────────────────────────────────────────────┘
│
▼
┌─────────────────────────────────────────────────────────────────┐
│ Claude becomes the Orchestrator │
│ ├── Reads state file │
│ ├── Decomposes goal into parallel tasks │
│ └── Spawns background agents via Task tool │
│ ├── Agent A: "Explore codebase" → wave-outputs/explore.md │
│ ├── Agent B: "Design schema" → wave-outputs/schema.md │
│ └── Agent C: "Setup project" → wave-outputs/setup.md │
└─────────────────────────────────────────────────────────────────┘
│
▼
┌─────────────────────────────────────────────────────────────────┐
│ Claude tries to exit/complete │
└─────────────────────────────────────────────────────────────────┘
│
▼
┌─────────────────────────────────────────────────────────────────┐
│ stop-hook.sh intercepts │
│ ├── Checks for <wave-complete>DONE</wave-complete> │
│ ├── If NOT complete → increment iteration, feed prompt back │
│ └── If complete → allow exit, cleanup │
└─────────────────────────────────────────────────────────────────┘
│
┌───────────────┴───────────────┐
▼ ▼
┌──────────────────────────┐ ┌──────────────────────────┐
│ Not Complete │ │ Complete │
│ ├── iteration++ │ │ ├── Remove state file │
│ └── Loop continues... │ │ └── Exit successfully │
└──────────────────────────┘ └──────────────────────────┘
/wave-orchestration:wave-preflight)
jq and perl are installed.claude/ directory is writablenpx claudepluginhub joshuarweaver/cascade-ai-ml-agents-misc-2 --plugin ferni-ai-wave-orchestrationIterative parallel orchestration combining Ralph Loop iteration with swarm agent parallelism. Spawns waves of agents, synthesizes results, and iterates until completion.
Advanced swarm orchestration patterns for research, development, testing, and complex distributed workflows
Context-optimized task orchestration for AI agents. Decomposes complex tasks into a DAG and dispatches focused sub-agents with minimal context prompts.
Multi-agent workflow orchestration via YAML. Ships the conductor skill so the assistant can validate, run, debug, and author workflow files for the conductor CLI.
Orchestrate complex workflows with DAG-based execution, parallel tasks, and run history tracking
Multi-phase subagent orchestration for Codex: 17 taskflow_* MCP tools (run/runs/resume/version/list/show/verify/compile/lint/peek/trace/replay/why_stale/recompute/reconcile_workspace/save/search) plus a routing skill. Background runs support status/wait/cancel; compile returns SVG + text outline. The server runs via npx (codex-taskflow).
Ultra-compressed communication mode. Cuts 65% of output tokens (measured) while keeping full technical accuracy by speaking like a caveman.