swarms todo

main
Kye 2 years ago
parent a8946335c4
commit 5db93f5249

@ -84,3 +84,8 @@ In the context of swarm LLMs, one could consider an **Omni-Vector Embedding Data
# Repos
[🐪CAMEL🐪](https://twitter.com/hwchase17/status/1645834030519296000)
[MultiAgent](https://github.com/rumpfmax/Multi-GPT/blob/master/multigpt/multi_agent_manager.py)
[AutoGPT](https://github.com/Significant-Gravitas/Auto-GPT)

@ -134,8 +134,6 @@ Here is the detailed roadmap of our priorities and planned features for the near
5. **Task Completion and Evaluation Logic**: Include task completion logic with meta prompting, and evaluate task completion on a scale from 0.0 to 1.0.
6. **Baby AGI Setup**: Set up Baby AGI with the AutoGPT instance as a tool for enhanced capabilities.
7. **Ocean Integration**: Use the [Ocean](https://github.com/kyegomez/Ocean) vector database as the main embedding database for all the agents, both boss and worker.
8. **Improved Communication**: Develop a universal vector database that is only used when a task is completed in this format `[TASK][COMPLETED]`.

@ -530,3 +530,37 @@ def get_token_ids_for_choose_model(model_name):
res = list(set(res))
return res
################# END
################# MultiAgent
from autogpt.agent import Agent
from swarms.agents.swarms import worker_node
class MultiAgent(worker_node):
def __init__(
self,
ai_name,
memory,
full_message_history,
prompt,
user_input,
agent_id
):
super().__init__(
ai_name=ai_name,
memory=memory,
full_message_history=full_message_history,
next_action_count=0,
prompt=prompt,
user_input=user_input,
)
self.agent_id = agent_id
self.auditory_buffer = [] # contains the non processed parts of the conversation
def receive_message(self, speaker, message):
self.auditory_buffer.append((speaker.ai_name, message))
Loading…
Cancel
Save