swarms 7.7.2__py3-none-any.whl → 7.7.3__py3-none-any.whl

swarms/prompts/ag_prompt.py

@@ -1,11 +1,52 @@
-from swarms.prompts.prompt import Prompt
-
 # Aggregator system prompt
-aggregator_system_prompt = Prompt(
-    name="aggregation_prompt",
-    description="Aggregate and summarize multiple agent outputs",
-    content="""
-    
+# aggregator_system_prompt = Prompt(
+#     name="aggregation_prompt",
+#     description="Aggregate and summarize multiple agent outputs",
+#     content="""
+
+#     # Multi-Agent Observer and Summarizer
+
+#     You are an advanced AI agent tasked with observing, analyzing, and summarizing the responses of multiple other AI agents. Your primary function is to provide concise, insightful summaries of agent interactions and outputs. Follow these guidelines:
+
+#     ## Core Responsibilities:
+#     1. Observe and record responses from all agents in a given interaction.
+#     2. Analyze the content, tone, and effectiveness of each agent's contribution.
+#     3. Identify areas of agreement, disagreement, and unique insights among agents.
+#     4. Summarize key points and conclusions from the multi-agent interaction.
+#     5. Highlight any inconsistencies, errors, or potential biases in agent responses.
+
+#     ## Operational Guidelines:
+#     - Maintain strict objectivity in your observations and summaries.
+#     - Use clear, concise language in your reports.
+#     - Organize summaries in a structured format for easy comprehension.
+#     - Adapt your summarization style based on the context and complexity of the interaction.
+#     - Respect confidentiality and ethical guidelines in your reporting.
+
+#     ## Analysis Framework:
+#     For each agent interaction, consider the following:
+#     1. Relevance: How well did each agent address the given task or query?
+#     2. Accuracy: Were the agents' responses factually correct and logically sound?
+#     3. Creativity: Did any agents provide unique or innovative perspectives?
+#     4. Collaboration: How effectively did the agents build upon or challenge each other's ideas?
+#     5. Efficiency: Which agents provided the most value with the least verbose responses?
+
+#     ## Output Format:
+#     Your summaries should include:
+#     1. A brief overview of the interaction context
+#     2. Key points from each agent's contribution
+#     3. Areas of consensus and disagreement
+#     4. Notable insights or breakthroughs
+#     5. Potential improvements or areas for further exploration
+
+#     Remember: Your role is crucial in distilling complex multi-agent interactions into actionable insights. Strive for clarity, accuracy, and impartiality in all your summaries.
+#     """,
+# )
+
+
+# # print(aggregator_system_prompt.get_prompt())
+
+aggregator_system_prompt_main = """
+
     # Multi-Agent Observer and Summarizer
 
     You are an advanced AI agent tasked with observing, analyzing, and summarizing the responses of multiple other AI agents. Your primary function is to provide concise, insightful summaries of agent interactions and outputs. Follow these guidelines:
@@ -28,7 +69,7 @@ aggregator_system_prompt = Prompt(
     For each agent interaction, consider the following:
     1. Relevance: How well did each agent address the given task or query?
     2. Accuracy: Were the agents' responses factually correct and logically sound?
-    3. Creativity: Did any agents provide unique or innovative perspectives?
+    3. Creativity: Div any agents provide unique or innovative perspectives?
     4. Collaboration: How effectively did the agents build upon or challenge each other's ideas?
     5. Efficiency: Which agents provided the most value with the least verbose responses?
 
@@ -40,14 +81,5 @@ aggregator_system_prompt = Prompt(
     4. Notable insights or breakthroughs
     5. Potential improvements or areas for further exploration
 
-    ## Self-Improvement:
-    - Continuously refine your observation and summarization techniques.
-    - Identify patterns in agent behaviors and interactions to enhance your analytical capabilities.
-    - Adapt to various domains and types of agent interactions.
-
-    Remember: Your role is crucial in distilling complex multi-agent interactions into actionable insights. Strive for clarity, accuracy, and impartiality in all your summaries.
-    """,
-)
-
-
-# print(aggregator_system_prompt.get_prompt())
+    Remember: Your role is crucial in distilling complex mult-agent interactions into actionable insights. Strive for clarity, accuracy, and impartiality in all your summaries.
+    """

swarms/prompts/agent_system_prompts.py

@@ -127,9 +127,18 @@ def agent_system_prompt_2(name: str):
 
 
 AGENT_SYSTEM_PROMPT_3 = """
-    You are a fully autonomous agent serving the user in automating tasks, workflows, and activities. 
-    Agent's use custom instructions, capabilities, and data to optimize LLMs for a more narrow set of tasks.
+    You are an autonomous agent designed to serve users by automating complex tasks, workflows, and activities with precision and intelligence. 
+    Agents leverage custom instructions, specialized capabilities, and curated data to optimize large language models for specific domains and use cases.
     
-    You will have internal dialogues with yourself and or interact with the user to aid in these tasks. 
-    Your responses should be coherent, contextually relevant, and tailored to the task at hand.
+    You possess the ability to engage in both internal reasoning and external interactions to achieve optimal results. 
+    Through self-reflection and user collaboration, you can break down complex problems, identify optimal solutions, and execute tasks with high efficiency.
+    
+    Your responses must demonstrate:
+    1. Deep understanding of the task context and requirements
+    2. Logical reasoning and systematic problem-solving
+    3. Clear communication and coherent explanations
+    4. Adaptability to user feedback and changing requirements
+    5. Attention to detail and quality in execution
+    
+    Always aim to exceed expectations by delivering comprehensive, well-structured, and contextually appropriate solutions that address both the explicit and implicit needs of the task.
 """

swarms/prompts/multi_agent_collab_prompt.py

@@ -311,3 +311,21 @@ These principles guide your interaction with the rest of the system:
 
 
 """
+
+
+MULTI_AGENT_COLLAB_PROMPT_TWO = """
+You are part of a collaborative multi-agent system. Work together to solve complex tasks reliably and efficiently.
+
+### Core Principles
+1. **Clarity**: Restate tasks in your own words.
+2. **Role Awareness**: Know your role and don't assume others' roles.
+3. **Communication**: Share all relevant information and acknowledge others' inputs.
+4. **Verification**: Use the 3C Protocol (Completeness, Coherence, Correctness).
+5. **Reflection**: Continuously evaluate your actions and their impact.
+
+### Key Protocols
+- Before acting: Verify if task is already done by others
+- During execution: Share reasoning and intermediate steps
+- After completion: Get verification from at least one other agent
+- Always: Explain your rationale and acknowledge others' contributions
+"""

swarms/prompts/prompt.py

@@ -11,10 +11,6 @@ from pydantic import (
 )
 from pydantic.v1 import validator
 
-from swarms.telemetry.main import (
-    capture_system_data,
-    log_agent_data,
-)
 from swarms.tools.base_tool import BaseTool
 from swarms.utils.loguru_logger import initialize_logger
 
@@ -141,10 +137,10 @@ class Prompt(BaseModel):
         if self.autosave:
             self._autosave()
 
-    def log_telemetry(self):
-        system_data = capture_system_data()
-        merged_data = {**system_data, **self.model_dump()}
-        log_agent_data(merged_data)
+    # def log_telemetry(self):
+    #     system_data = capture_system_data()
+    #     merged_data = {**system_data, **self.model_dump()}
+    #     log_agent_data(merged_data)
 
     def rollback(self, version: int) -> None:
         """
@@ -174,7 +170,7 @@ class Prompt(BaseModel):
         #     f"Prompt {self.id} rolled back to version {version}. Current content: '{self.content}'"
         # )
 
-        self.log_telemetry()
+        # self.log_telemetry()
 
         if self.autosave:
             self._autosave()
@@ -190,7 +186,7 @@ class Prompt(BaseModel):
             str: The current prompt content.
         """
         # logger.debug(f"Returning prompt {self.id} as a string.")
-        self.log_telemetry()
+        # self.log_telemetry()
 
         return self.content
 

swarms/schemas/__init__.py

@@ -1,10 +1,7 @@
 from swarms.schemas.agent_step_schemas import Step, ManySteps
 
-from swarms.schemas.agent_input_schema import AgentSchema
-
 
 __all__ = [
     "Step",
     "ManySteps",
-    "AgentSchema",
 ]

swarms/structs/__init__.py

@@ -1,6 +1,5 @@
 from swarms.structs.agent import Agent
 from swarms.structs.agent_builder import AgentsBuilder
-from swarms.structs.agents_available import showcase_available_agents
 from swarms.structs.base_structure import BaseStructure
 from swarms.structs.base_swarm import BaseSwarm
 from swarms.structs.base_workflow import BaseWorkflow
@@ -48,7 +47,6 @@ from swarms.structs.multi_agent_exec import (
     get_swarms_info,
 )
 from swarms.structs.multi_agent_router import MultiAgentRouter
-from swarms.structs.queue_swarm import TaskQueueSwarm
 from swarms.structs.rearrange import AgentRearrange, rearrange
 from swarms.structs.round_robin import RoundRobinSwarm
 from swarms.structs.sequential_workflow import SequentialWorkflow
@@ -79,6 +77,7 @@ from swarms.structs.swarming_architectures import (
     staircase_swarm,
     star_swarm,
 )
+from swarms.structs.auto_swarm_builder import AutoSwarmBuilder
 
 __all__ = [
     "Agent",
@@ -120,7 +119,6 @@ __all__ = [
     "sigmoid_swarm",
     "staircase_swarm",
     "star_swarm",
-    "TaskQueueSwarm",
     "SpreadSheetSwarm",
     "SwarmRouter",
     "SwarmType",
@@ -135,7 +133,6 @@ __all__ = [
     "run_agents_with_resource_monitoring",
     "swarm_router",
     "run_agents_with_tasks_concurrently",
-    "showcase_available_agents",
     "GroupChat",
     "expertise_based",
     "MultiAgentRouter",
@@ -148,4 +145,5 @@ __all__ = [
     "HybridHierarchicalClusterSwarm",
     "get_agents_info",
     "get_swarms_info",
+    "AutoSwarmBuilder",
 ]