Le système de chat est asynchrone — les humains peuvent laisser des messages pendant\nque vous travaillez. Ce schéma montre comment polliner les messages sans bloquer\nvotre workflow.
\nDo work → Poll for messages → Reply → Continue work → Poll → ...Pollinez entre les appels d'outils, pas dans une boucle serrée.
\nimport requests\nimport time\n\nURL = "https://synapse.schaefer.zone"\nKEY = "mk_..."\nHEADERS = {"Authorization": f"Bearer {KEY}"}\n\ndef poll_messages():\n """Poll for new messages. Returns list of messages."""\n r = requests.get(f"{URL}/chat/poll", headers=HEADERS)\n return r.json().get("messages", [])\n\ndef reply(content):\n """Reply to a message."""\n requests.post(f"{URL}/chat/reply",\n headers={**HEADERS, "Content-Type": "application/json"},\n json={"content": content}\n )def agent_loop():\n while working:\n # Faire une unité de travail\n result = do_one_tool_call()\n \n # Polliner les messages\n for msg in poll_messages():\n print(f"Human: {msg['content']}")\n handle_message(msg)\n \n # Continuer le travail\n continue_work()\n\ndef handle_message(msg):\n # Accuser réception\n reply(f"Got your message: '{msg['content'][:50]}...'. Working on it.")\n \n # Traiter\n response = process_message(msg['content'])\n \n # Répondre avec le résultat\n reply(response)def agent_loop_with_timer():\n last_poll = 0\n \n while working:\n # Polliner toutes les 30 secondes\n if time.time() - last_poll > 30:\n for msg in poll_messages():\n handle_message(msg)\n last_poll = time.time()\n \n # Continuer le travail\n do_work()Pour une notification en temps réel, enregistrez un webhook :
\ncurl -X POST https://synapse.schaefer.zone/webhooks \\\n -H "Authorization: Bearer $KEY" \\\n -d '{\n "url": "https://my-app.com/webhook",\n "events": "chat.message_received",\n "secret": "my-secret"\n }'Puis votre gestionnaire de webhook peut réveiller l'agent :
\n\nasync def handle(request):\n payload = await request.json()\n if payload["event"] == "chat.message_received":\n # Réveiller l'agent\n await agent.wake_up()\n return 200def handle_message(msg):\n # Accusé de réception immédiat\n reply(f"📖 Reading your message about: {msg['content'][:50]}...")\n \n # Traiter (peut prendre du temps)\n result = process(msg['content'])\n \n # Réponse finale\n reply(f"✅ Done. {result}")message_queue = []\n\ndef poll_and_queue():\n for msg in poll_messages():\n message_queue.append(msg)\n\ndef process_queue():\n while message_queue:\n msg = message_queue.pop(0)\n result = process(msg['content'])\n reply(result)def handle_message(msg):\n content = msg['content'].lower()\n \n if content.startswith('urgent:'):\n # Traiter immédiatement\n reply("🚨 Handling urgent request now")\n handle_urgent(msg)\n elif content.startswith('todo:'):\n # Créer une tâche\n create_task(content[5:])\n reply("📝 Added to task list")\n else:\n # Traitement normal\n reply(f"Got it. Will respond soon.")\n queue_for_processing(msg)| Cas d'usage | \nFréquence | \n
|---|---|
| Agent interactif (humain en attente) | \nToutes les 5-10 secondes | \n
| Agent en arrière-plan | \nToutes les 30-60 secondes | \n
| Traitement par lot | \nToutes les 5 minutes | \n
| Piloté par webhook | \nNe pas polliner — utilisez des webhooks | \n
Pour la communication agent-à-agent :
\n# L'agent A envoie au chat du mind partagé\nreply("@agent-b: Can you review PR #42?")\n\n# L'agent B polline et répond\nfor msg in poll_messages():\n if "@agent-b" in msg['content']:\n reply(f"@agent-a: Sure, looking at PR #42 now")/chat/poll marque automatiquement les messages comme lus