Add FAQ

docs_new/faq.mdx

@@ -8,28 +8,77 @@ Find answers to common questions about Agency Swarm.
 
 <AccordionGroup defaultOpen={true}>
 
-<Accordion title="What is Agency Swarm?" icon="bee" iconType="solid">
-**A**: Agency Swarm is an open-source framework designed to facilitate the creation and orchestration of multi-agent AI systems.
+<Accordion title="How can I restart or continue an agency creation run with Genesis after encountering an error?" icon="rotate">
+**A**: You can implement thread callbacks to save and load thread IDs in a local file. This allows you to maintain conversation context even after interruptions. Here's the basic approach:
+
+- Create load and save callback functions
+- Use these callbacks in your `GenesisAgency` initialization
+- Control threads via a `chat_id` parameter
 </Accordion>
 
-<Accordion title="How do I install Agency Swarm?" icon="download" iconType="solid">
-**A**: Install via pip using:
+<Accordion title="How do I implement thread callbacks for conversation continuity?" icon="messages">
+**A**: You can use threads callbacks to save thread ids in a local file:
+
+Create load function:
+
+```python
+def load_threads(chat_id):
+    if os.path.exists(f"{chat_id}_threads.json"):
+        with open(f"{chat_id}_threads.json", "r") as file:
+            return json.load(file)
+    return []
+```
+
+Create save function:
 
-```bash
-pip install agency-swarm
+```python
+def save_threads(new_threads, chat_id):
+    with open(f"{chat_id}_threads.json", "w") as file:
+        json.dump(new_threads, file)
 ```
+Implement in agency:
+
+```python
+agency = GenesisAgency(
+    with_browsing=False,
+    threads_callbacks={
+        'load': lambda: load_threads(chat_id=chat_id),
+        'save': lambda new_threads: save_threads(new_threads, chat_id=chat_id)
+    })
+```
+Note: These code examples are simplified for illustration. Production implementations should include proper error handling, security measures, and scalability considerations.
 </Accordion>
 
-<Accordion title="Can I use custom models with Agency Swarm?" icon="gear" iconType="solid">
-**A**: Yes, you can integrate open-source models or custom language models that mimic the OpenAI Assistants API.
+<Accordion title="Should I use local file storage for thread management in a production environment?" icon="rocket">
+**A**: No, for production environments, especially with multiple users or stateless containers, you should:
+
+- Use a database instead of local file storage
+- Implement proper state management
+- Consider scalability requirements
+- Handle concurrent users appropriately
 </Accordion>
 
-<Accordion title="How do agents communicate within an agency?" icon="comments" iconType="solid">
-**A**: Agents communicate using the `SendMessage` tool, with communication flows defined in the `agency_chart` parameter of the `Agency` class.
+<Accordion title="Will previous messages be visible in Gradio when continuing a conversation?" icon="comments">
+**A**: No, previous messages won't be displayed in the Gradio interface. However, the agents will retain the context of these messages and continue the conversation appropriately.
 </Accordion>
 
-<Accordion title="Where can I find examples and tutorials?" icon="book" iconType="solid">
-**A**: Examples are available in the [agency-swarm-lab](https://github.com/VRSEN/agency-swarm-lab) repository and on the [YouTube channel](https://youtube.com/@vrsen).
+<Accordion title="How can I serve an Agency as an API using FastAPI?" icon="book">
+**A**: Here's an example of how to serve an Agency as an API using FastAPI:
+
+- Define the agency within the chat endpoint:
+```python
+@app.post("/chat")
+async def chat(user_request: UserRequest):
+    chat_id = user_request.chat_id or str(uuid4())
+    agency = Agency([...],
+        threads_callbacks={
+            'load': lambda: load_threads(chat_id=chat_id),
+            'save': lambda new_threads: save_threads(new_threads, chat_id=chat_id)
+        })
+    response = agency.get_completion(user_request.message)
+    return {"chat_id": chat_id, "response": response}
+```
+
 </Accordion>
 
 </AccordionGroup>