🌊 The original agent harness. Deploy intelligent multi-player swarms, coordinate autonomous workflows, and build conversational AI systems. Features adaptive memory, self-learning intelligence, federation, vector RAG integration, and native Claude Code / Codex / Hermes and many more Integrated
TypeScript agentic-aiagentic-frameworkagentic-workflowagentsai-agentsai-assistantai-skillsautonomous-agentsclaudecodexdsh-pluginharnessmcp-servermulti-agentmulti-agent-systemsnpmskillsswarmswarm-intelligencetypescript
Checkpoint: File edits
rUve43da122b18e
diff --git a/.claude/agents/dual-mode/dual-orchestrator.yaml b/.claude/agents/dual-mode/dual-orchestrator.yamlnew file mode 100644index 0000000000..5201f23b41--- /dev/null+++ b/.claude/agents/dual-mode/dual-orchestrator.yaml@@ -0,0 +1,175 @@+name: dual-orchestrator+description: Orchestrates Claude Code (interactive) + Codex (headless) for hybrid workflows+type: orchestrator+platform: dual++# Mode configuration+modes:+ interactive:+ platform: claude-code+ use_for:+ - complex-reasoning+ - architecture-decisions+ - debugging+ - real-time-review++ headless:+ platform: codex+ use_for:+ - parallel-execution+ - batch-processing+ - code-generation+ - documentation+ - testing++# Routing rules+routing:+ # Route to headless Codex+ codex_patterns:+ - "implement * in parallel"+ - "generate * files"+ - "write tests for *"+ - "document *"+ - "batch process *"++ # Keep in interactive Claude Code+ claude_patterns:+ - "explain *"+ - "debug *"+ - "design *"+ - "review with me *"+ - "help me understand *"++# MCP Integration (shared)+mcp:+ enabled: true+ shared_tools:+ - memory_search+ - memory_store+ - swarm_init+ - swarm_status+ - agent_spawn++# Workflow patterns+workflows:+ hybrid_development:+ name: Hybrid Development Flow+ description: Use Claude Code for design, Codex for implementation+ steps:+ - phase: design+ platform: claude-code+ interactive: true+ tasks:+ - Discuss requirements+ - Design architecture+ - Store design in memory++ - phase: implement+ platform: codex+ parallel: true+ workers:+ - type: coder+ count: 2+ - type: tester+ count: 1++ - phase: review+ platform: claude-code+ interactive: true+ tasks:+ - Review implementation+ - Discuss improvements+ - Finalize++ parallel_feature:+ name: Parallel Feature Implementation+ steps:+ - action: swarm_init+ args: { topology: hierarchical, maxAgents: 6 }++ - action: spawn_headless+ workers:+ - { role: architect, task: "Design feature" }+ - { role: coder-1, task: "Implement core" }+ - { role: coder-2, task: "Implement API" }+ - { role: tester, task: "Write tests" }+ - { role: docs, task: "Write documentation" }++ - action: wait_all++ - action: interactive_review+ platform: claude-code++# Prompts+prompts:+ decide_platform: |+ Analyzing task: {{task}}++ ROUTING DECISION:+ {{#if requires_interaction}}+ → Use Claude Code (interactive)+ Reason: {{interaction_reason}}+ {{else}}+ → Use Codex (headless)+ Reason: Can run in background, parallelizable+ {{/if}}++ spawn_hybrid: |+ HYBRID WORKFLOW: {{workflow_name}}++ INTERACTIVE (Claude Code):+ {{interactive_tasks}}++ HEADLESS (Codex - parallel):+ {{#each headless_workers}}+ claude -p "{{this.prompt}}" --session-id {{this.id}} &+ {{/each}}++ COORDINATION:+ All workers share memory via claude-flow MCP tools.+ Results stored in 'results' namespace.++ aggregate: |+ COLLECTING RESULTS:++ ```bash+ npx claude-flow memory list --namespace results+ ```++ SYNTHESIS:+ {{#each results}}+ - {{this.worker}}: {{this.summary}}+ {{/each}}++# Examples+examples:+ - name: Build API feature+ interactive_phase: |+ Let's design the API endpoints together.+ I'll help you think through the data models+ and error handling strategies.++ headless_phase: |+ claude -p "Implement GET /users endpoint" &+ claude -p "Implement POST /users endpoint" &+ claude -p "Write integration tests" &+ wait++ review_phase: |+ Now let's review what the workers produced.+ I'll help identify any issues or improvements.++ - name: Refactor module+ interactive_phase: |+ Let's analyze the current code structure+ and plan the refactoring approach.++ headless_phase: |+ claude -p "Refactor authentication module" &+ claude -p "Update all imports" &+ claude -p "Run test suite" &+ wait++ review_phase: |+ Let's verify the refactoring is complete+ and all tests pass.