┌─────────────────────────────────────────┐
│              Orchestrator               │
│  (Assign tasks, monitor, coordinate)    │
└────────────────┬────────────────────────┘
                 │
    ┌────────────┼────────────┐
    ▼            ▼            ▼
┌────────┐  ┌────────┐  ┌────────┐
│ Agent 1│  │ Agent 2│  │ Agent 3│
│ (Task) │  │ (Task) │  │ (Task) │
└────┬───┘  └────┬───┘  └────┬───┘
     │           │           │
     └───────────┴───────────┘
                 │
                 ▼
        ┌────────────────┐
        │ Knowledge Base │
        │   (ChromaDB)   │
        └────────────────┘

def spawn_agent(agent_id: str, task: str):
    # Create PTY for agent terminal
    master, slave = pty.openpty()
    
    # Spawn process
    process = subprocess.Popen(
        ['claude', '--task', task],
        stdin=slave,
        stdout=slave,
        stderr=slave,
        start_new_session=True
    )
    
    return {
        'id': agent_id,
        'process': process,
        'master_fd': master,
        'status': 'running'
    }

def monitor_agent(agent):
    # Read output non-blocking
    ready, _, _ = select.select([agent['master_fd']], [], [], 0.1)
    if ready:
        output = os.read(agent['master_fd'], 4096).decode()
        return output
    return None

def harvest_knowledge(completed_task):
    # Extract learnings
    learnings = {
        'task': completed_task['description'],
        'solution': completed_task['output'],
        'patterns': extract_patterns(completed_task['output']),
        'timestamp': datetime.now().isoformat()
    }
    
    # Store in vector DB
    collection.add(
        documents=[learnings['solution']],
        metadatas=[learnings],
        ids=[f"learning_{uuid.uuid4()}"]
    )

import chromadb

client = chromadb.Client()
collection = client.get_or_create_collection("knowledge_base")

collection.add(
    documents=["Solution text here"],
    metadatas=[{"source": "agent_1", "task": "debug"}],
    ids=["unique_id"]
)

results = collection.query(
    query_texts=["How to fix null pointer?"],
    n_results=5
)

┌──────────────┐
│ Agent runs   │
│    task      │
└──────┬───────┘
       ▼
┌──────────────┐
│ Task result  │
│  extracted   │
└──────┬───────┘
       ▼
┌──────────────┐
│  Knowledge   │  ← Store patterns, solutions
│  harvested   │
└──────┬───────┘
       ▼
┌──────────────┐
│  Next agent  │  ← Query relevant context
│ uses context │
└──────────────┘

Multi-agent task?
├── Need new agent? → spawn_agent()
├── Agent stuck? → Check PTY buffer, restart if needed
├── Task complete? → Harvest knowledge → ChromaDB
├── Similar task? → Query ChromaDB for context
└── Coordination? → Use message queue/shared state