Role-Playing Scraper for Report & Knowledge Graph Generation#
You can also check this cookbook in colab here
This notebook demonstrates how to set up and leverage CAMELโs Retrieval-Augmented Generation (RAG) combined with Firecrawl for efficient web scraping, multi-agent role-playing tasks, and knowledge graph construction. We will walk through an example of conducting a comprehensive study of the Turkish shooter in the 2024 Paris Olympics by using Mistralโs models.
In this notebook, youโll explore:
CAMEL: A powerful multi-agent framework that enables Retrieval-Augmented Generation and multi-agent role-playing scenarios, allowing for sophisticated AI-driven tasks.
Mistral: Utilized for its state-of-the-art language models, which enable tool-calling capabilities to execute external functions, while its powerful embeddings are employed for semantic search and content retrieval.
Firecrawl: A robust web scraping tool that simplifies extracting and cleaning content from various web pages.
AgentOps: Track and analysis the running of CAMEL Agents.
Qdrant: An efficient vector storage system used with CAMELโs AutoRetriever to store and retrieve relevant information based on vector similarities.
Neo4j: A leading graph database management system used for constructing and storing knowledge graphs, enabling complex relationships between entities to be mapped and queried efficiently.
DuckDuckGo Search: Utilized within the SearchToolkit to gather relevant URLs and information from the web, serving as the primary search engine for retrieving initial content.
Unstructured IO: Used for content chunking, facilitating the management of unstructured data for more efficient processing.
This setup not only demonstrates a practical application but also serves as a flexible framework that can be adapted for various scenarios requiring advanced web information retrieval, AI collaboration, and multi-source data aggregation.
โญ Star the Repo
If you find CAMEL useful or interesting, please consider giving it a star on our CAMEL GitHub Repo! Your stars help others find this project and motivate us to continue improving it.
๐ฆ Installation#
First, install the CAMEL package with all its dependencies:
[ ]:
%pip install "camel-ai[all]==0.2.1"
๐ Setting Up API Keys#
Youโll need to set up your API keys for Mistral AI, Firecrawl and AgentOps. This ensures that the tools can interact with external services securely.
You can go to here to get free API Key from AgentOps
[2]:
import os
from getpass import getpass
# Prompt for the AgentOps API key securely
agentops_api_key = getpass('Enter your API key: ')
os.environ["AGENTOPS_API_KEY"] = agentops_api_key
Enter your API key: ยทยทยทยทยทยทยทยทยทยท
Your can go to here to get API Key from Mistral AI with free credits.
[3]:
# Prompt for the API key securely
mistral_api_key = getpass('Enter your API key: ')
os.environ["MISTRAL_API_KEY"] = mistral_api_key
Enter your API key: ยทยทยทยทยทยทยทยทยทยท
Set up the Mistral Large 2 model using the CAMEL ModelFactory. You can also configure other models as needed.
[4]:
from camel.models import ModelFactory
from camel.types import ModelPlatformType, ModelType
from camel.configs import MistralConfig
# Set up model
mistral_large_2 = ModelFactory.create(
model_platform=ModelPlatformType.MISTRAL,
model_type=ModelType.MISTRAL_LARGE,
model_config_dict=MistralConfig(temperature=0.2).as_dict(),
)
Your can go to here to get API Key from Firecrawl with free credits.
[5]:
# Prompt for the Firecrawl API key securely
firecrawl_api_key = getpass('Enter your API key: ')
os.environ["FIRECRAWL_API_KEY"] = firecrawl_api_key
Enter your API key: ยทยทยทยทยทยทยทยทยทยท
๐ Web Scraping with Firecrawl#
Firecrawl is a powerful tool that simplifies web scraping and cleaning content from web pages. In this section, we will scrape content from a specific post on the CAMEL AI website as an example.
[13]:
from camel.loaders import Firecrawl
firecrawl = Firecrawl()
# Scrape and clean content from a specified URL
response = firecrawl.scrape(
url="https://www.camel-ai.org/post/crab"
)
print(response["markdown"])
9th August, 2024
# CRAB: Cross-environment Agent Benchmark for Multimodal Language Model Agents
> Abstract: Recently, spearheaded by the CAMEL-AI community, a pioneer in open-source multi-agent projects, researchers from institutions such as King Abdullah University of Science and Technology, Oxford University, University of Tokyo, Carnegie Mellon University, Stanford University, and Tsinghua University have developed a cross-platform multimodal agent benchmark framework: CRAB, innovatively enabling agents to operate multiple devices simultaneously.
### Introduction
With the rapid development of multimodal large language models (MLLM), many agents capable of operating graphical user interfaces (GUIs) have emerged this year. Various companies have launched their innovative solutions, creating intense competition. GUI agents, leveraging powerful visual understanding and reasoning abilities of large models, can now efficiently and flexibly complete tasks such as booking appointments, shopping, and controlling smart homes.
**This raises the question: will future agents truly be able to sit in front of a computer and work on my behalf?**
However, in today's era of the Internet of Everything, most work requires the coordination of multiple devices. For example, taking a photo with a phone and then transferring it to a computer for editing involves crossing two different devices (environments). Currently, these GUI agents can only operate on a single device, making what is an easy task for humans exceedingly difficult for today's agents.
Researchers from the CAMEL-AI community noticed this problem and proposed the first cross-environment, multi-device agent benchmark frameworkโCRAB, the **CR** oss-environment **A** gent **B** enchmark.
Paper link: [https://arxiv.org/abs/2407.01511](https://arxiv.org/abs/2407.01511)
The CAMEL framework ( [https://github.com/camel-ai](https://github.com/camel-ai)) developed by the CAMEL-AI community is one of the earliest open-source multi-agent projects based on large language models. Therefore, community members are researchers and engineers with rich research and practical experience in the field of agents.
In CRAB, the authors not only designed a network-based multi-environment architecture that enables agents to operate multiple devices simultaneously to complete tasks, but also proposed two new technologies to address the issues existing in current agent benchmarks: the graph evaluator and task synthesis. CRAB is not only a brand new benchmark tool but also provides an interaction protocol and its implementation between the environment and agents, which is expected to become an important foundation for agents in practical fields.
The authors believe that CRAB will become one of the standards for evaluating GUI agents in the future, and thus have put considerable effort into improving the framework's usability. The entire codebase adopts a modular design, with the configuration of each environment abstracted into independent and reusable components. Users can easily and quickly build multiple custom environments like building blocks and create their own benchmarks based on them.
For users who wish to evaluate their agents' performance using CRAB, the authors thoughtfully provide a hard disk image on the Google Cloud Platform. With just one click, all the tedious configurations of virtual machines, deep learning models, Python packages, etc., will be completed automatically, allowing users to immediately engage in important experiments.
Currently, CRAB's paper has been published on Arxiv, and the related code and data are open-sourced on the CAMEL-AI community's GitHub.
GitHub repository: [https://github.com/camel-ai/crab](https://github.com/camel-ai/crab)
What does CRAB look like in operation? Let's take a look at the video below:
โ
The process of testing a multi-agent system on CRAB.
First, the system extracts tasks from the dataset, passes the task instructions to the main agent, and initializes the corresponding graph evaluator in CRAB.
The workflow is a loop: the main agent observes, plans, and instructs the sub-agents; each sub-agent corresponds to an environment. In the diagram, two sub-agents are responsible for the Ubuntu system computer and the Android system phone, respectively, and the sub-agents perform operations in their respective platforms.
The graph evaluator monitors the state of each environment in the platform, updates the progress of the agent in completing the task after each operation, and outputs evaluation metrics.
Having understood how CRAB works, it's time to see how current models perform on this new benchmark. The authors introduced CRAB-Benchmark-v0, a dataset supporting two environments with a total of 100 tasks, and tested several state-of-the-art models.
As shown, the best-performing GPT-4o scored only 35.26 (CR refers to completion rate).
Cross-platform tasks are much more complex than single-platform tasks, and achieving the top score already demonstrates the outstanding ability of the GPT-4 series models in solving practical problems. However, we believe that emerging new methods and models can achieve better scores on CRAB, truly becoming efficient tools for solving real-world problems.
### Cross-Platform Multimodal Agent Evaluation
CRAB provides a comprehensive interactive agent evaluation framework. Through CRAB's foundational setup, agents can operate on various devices and platforms simultaneously, efficiently completing tasks in multiple isolated systems.
The authors propose a new evaluation method called the graph evaluator, which differs from traditional methods based on final goals or action trajectories. The graph evaluator checks the intermediate states of task completion by breaking down tasks into multiple sub-goals.
Each sub-goal is assigned an evaluation function to verify its completion, with each sub-goal being treated as a node in the graph evaluator.
The graph structure describes the precedence and parallel relationships between sub-goals, thus providing fine-grained evaluation metrics; at the same time, the independent evaluation function for each sub-goal inherently adapts to multiple platforms.
The table above compares CRAB with existing frameworks, including several key capabilities involved in testing:
- **Interactive Environment:** Indicates whether an interactive platform or static dataset is used.
- **Multimodal Observation:** Indicates whether multimodal input (e.g., screenshots) is supported.
- **Cross-platform:** Indicates whether multiple operating systems or platforms are supported simultaneously.
- **Evaluation:** Describes evaluation metrics, divided into goal-based (only checking if the final goal is completed), trajectory-based (comparing the agent's action trajectory with a predefined standard action sequence), multiple (varies by task), or graph-based (each node as an intermediate checkpoint in a DAG).
- **Task Construction:** Shows the method of constructing tasks in the test dataset, including manually created, LLM-inspired (e.g., LLM generates task drafts but is verified and annotated by humans), template (multiple tasks generated based on manually written templates), or sub-task composition (composing multiple sub-tasks to construct task descriptions and evaluators).
Based on the CRAB framework, the authors developed a benchmark dataset, CRAB Benchmark v0, supporting Android and Ubuntu environments.
The benchmark includes 100 real-world tasks, covering various levels of difficulty for cross-platform and single-platform tasks. Tasks involve a variety of common issues and use multiple practical applications and tools, including but not limited to calendars, emails, maps, web browsers, and terminals, also replicating common coordination methods between smartphones and computers.
### Problem Definition
Assume an agent autonomously executes tasks on a device (such as a desktop). The device is usually equipped with input devices (like a mouse and keyboard) and output devices (like a screen) for human-computer interaction. The authors refer to a device or application with a fixed input method and output method as an environment.
Formally, a single environment can be defined as a reward-free partially observable Markov decision process (Reward-free POMDP), represented by a tuple M := (S, A, T, O), where S represents the state space, A represents the action space, T: S ร A โ S is the transition function, and O is the observation space.
Considering the collaborative nature of multiple devices in real-world scenarios, multiple platforms can be combined into a set M = {M1, M2, ..., Mn}, where n is the number of platforms, and each platform can be represented as Mj = (Sj, Aj, Tj, Oj).
A task requiring operation across multiple platforms can be formalized as a tuple (M, I, R), where M is the platform set, I is the task goal described in natural language, and R is the task reward function.
The authors call the algorithm responsible for completing the task the agent system. Each agent in the agent system uses a fixed backend model and predefined system prompt, retaining its dialogue history. The agent system can be a single agent or a multi-agent system with multiple agents cooperating.
### Graph of Decomposed Tasks
Breaking down complex tasks into simpler sub-tasks is an effective method for prompting large language models. The authors introduced this concept into the benchmark field, breaking down complex tasks into sub-tasks with precedence and parallel relationships, known as the Graph of Decomposed Tasks (GDT) as shown above.
GDT uses a graph-based task decomposition method: using a DAG structure to represent decomposed sub-tasks.
In GDT, each node is a sub-task, formalized as a tuple (m, i, r), where m specifies the environment for executing the sub-task, i provides natural language instructions, and r represents the reward function. The reward function evaluates the state of environment m and outputs a boolean value to determine if the sub-task is completed. Edges in GDT represent the precedence relationships between sub-tasks.
### Graph Evaluator
To evaluate the capabilities of large language models as agents, most benchmarks rely solely on the final state of the platform after the agent's operations.
Simply judging whether the final goal is successful or failed is obviously not fair; it's like in a math exam, even if you can't solve the big problem, you should get points for writing some solution steps.
Another method is trajectory-based matching, comparing the agent's operations with a predefined standard operation sequence (Label) for each task.
However, in real-world systems, tasks may have multiple valid execution paths. For example, copying a file can be done using a file manager or a command line. Specifying a unique correct path is unfair to agents that achieve the goal in different ways.
Therefore, this paper adopts a graph evaluator synchronized with platform states, tracking the agent's progress through the current state of sub-task completion.
In addition to the traditional success rate (SR), which marks a task as successful only when all sub-tasks are completed, the authors introduced three metrics to measure the agent's performance and efficiency:
1. **Completion Rate (CR):** Measures the proportion of completed sub-task nodes, calculated as the number of completed nodes/total nodes. This metric intuitively reflects the agent's progress on the given task.
2. **Execution Efficiency (EE):** Calculated as CR/A, where A represents the number of actions performed, reflecting the agent's task execution efficiency.
3. **Cost Efficiency (CE):** Calculated as CR/T, where T is the total number of tokens used by the agent, evaluating the agent's efficiency in terms of cost.
### Experiments
To run in Crab Benchmark-v0, the backend model needs to support the following features:
1. Support multimodal mixed input: The system provides both screenshots and text instructions as prompts.
2. Support multi-turn conversations: All tasks require the agent to perform multiple operations, so historical messages must be stored in context.
3. Generate structured output through Function Call or similar functions: Used to execute operations in the environment.
The experiments selected four multimodal models that meet these criteria: GPT-4o, GPT-4 Turbo, Gemini 1.5 Pro, and Claude 3 Opus.
To compare the performance differences between multi-agent and single-agent systems, the paper designed three different agent systems:
1. **Single Agent:** A single agent handles the entire process from understanding the task, observing the environment, planning, to executing actions.
2. **Multi-agent by Functionality:** Consists of a main agent and a sub-agent. The main agent observes the environment and provides instructions to the sub-agent, which translates the instructions into specific operations.
3. **Multi-agent by Environment:** Consists of a main agent and multiple sub-agents. Each sub-agent is responsible for one environment. The main agent understands the task, plans the execution process, and provides instructions to each sub-agent. The sub-agents observe their respective environments and translate the instructions into specific operations.
The combinations of different agent systems and backend models provide multiple dimensions for comparison. Additionally, the paper compares the performance of models in tasks involving different platforms:
**Ubuntu single-platform tasks:**
**Android single-platform tasks:**
**Cross-platform tasks:**
Through data analysis, the paper draws several conclusions:
**Performance Differences Among Models:**
1. GPT-4o has the highest success and completion rates overall.
2. GPT-4 TURBO performs better in cost efficiency (CE) than other models.
3. Gemini 1.5 Pro and Claude 3 Opus struggle with task completion, finding it almost impossible to complete tasks. **โ**
**Efficiency Metrics Reflect Different Characteristics of Models:**
1. GPT-4 TURBO shows excellent cost efficiency in the single-agent mode, demonstrating cost-effective performance.
2. GPT-4o maintains a balance between efficiency and performance, especially in the single-agent mode.
3. Gemini 1.5 Pro shows low efficiency and incomplete cost efficiency metrics, mainly due to its low completion rate.
**Evaluation of Termination Reasons Indicates Areas for Improvement:**
1. All models have a high percentage of reaching the step limit (RSL), indicating that agents often run out of steps without achieving the final goal.
2. Gemini 1.5 Pro has a high rate of invalid actions (IA), highlighting its inability to stably generate the correct format for interacting with the environment.
3. The false completion (FC) rate in multi-agent systems is higher than in single-agent systems, indicating that message loss during communication between multiple agents can easily cause the executing sub-agent to misjudge.
### ๐ซ Thanks from everyone at CAMEL-AI
Hello there, passionate AI enthusiasts! ๐ We are ๐ซ CAMEL-AI.org, a global coalition of students, researchers, and engineers dedicated to advancing the frontier of AI and fostering a harmonious relationship between agents and humans.
**๐ Our Mission:** To harness the potential of AI agents in crafting a brighter and more inclusive future for all. Every contribution we receive helps push the boundaries of whatโs possible in the AI realm.
**๐ Join Us:** If you believe in a world where AI and humanity coexist and thrive, then youโre in the right place. Your support can make a significant difference. Letโs build the AI society of tomorrow, together!
- Find all our updates on [X](https://twitter.com/CamelAIOrg).
- Make sure to star our [GitHub](https://github.com/camel-ai) repositories.
- Join our [Discord,](https://discord.gg/nCpraan3sS) [WeChat](https://ghli.org/camel/wechat.png) or [Slack](https://join.slack.com/t/camel-ai/shared_invite/zt-2icssxnkj-YHwFVhoZHMYpIG~ZU86WVw) community.
- You can contact us by email: camel.ai.team@gmail.com
- Dive deeper and explore our projects on [https://www.camel-ai.org/](https://www.camel-ai.org/)
> โ
Get started
## Build your multi-agent system today
Get started by joining our community
[Join Discord](https://discord.gg/rBnhgxy6Fg)
Finding the scaling law of agents
RESEARCH
[Agent Trust](https://agent-trust.camel-ai.org/) [CRAB](https://github.com/camel-ai/crab?tab=readme-ov-file)
RESOURCES
[Blog](/blog) [Docs](https://docs.camel-ai.org/) [Features & Integrations](/integration) [Team](/about)
SOCIALS
[Discord](https://discord.gg/3rWmpWegBK) [Github](https://github.com/camel-ai/camel) [LinkedIn](https://www.linkedin.com/company/camel-ai-org/) [Twitter](https://x.com/CamelAIOrg) [YouTube](https://www.youtube.com/@CamelAI)
Copyright ยฉ 2024 Eigent AI - All Rights Reserved.
```javascript
pip install camel-ai[all]==0.1.6.4
```
๐ Firecrawl makes obtaining clean, LLM-friendly content from URL effortless!
๐ ๏ธ Web Information Retrieval using CAMELโs RAG and Firecrawl#
In this section, weโll demonstrate how to retrieve relevant information from a list of URLs using CAMELโs RAG model. This is particularly useful for aggregating and analyzing data from multiple sources.
Setting Up Firecrawl with CAMELโs RAG#
The following function retrieves relevant information from a list of URLs based on a given query. It combines web scraping with Firecrawl and CAMELโs AutoRetriever for a seamless information retrieval process.
[8]:
from camel.retrievers import AutoRetriever
from camel.toolkits import OpenAIFunction, SearchToolkit
from camel.types import ModelPlatformType, ModelType, StorageType
from camel.embeddings import MistralEmbedding
[14]:
def retrieve_information_from_urls(urls: list[str], query: str) -> str:
r"""Retrieves relevant information from a list of URLs based on a given
query.
This function uses the `Firecrawl` tool to scrape content from the
provided URLs and then uses the `AutoRetriever` from CAMEL to retrieve the
most relevant information based on the query from the scraped content.
Args:
urls (list[str]): A list of URLs to scrape content from.
query (str): The query string to search for relevant information.
Returns:
str: The most relevant information retrieved based on the query.
Example:
>>> urls = ["https://example.com/article1", "https://example.com/
article2"]
>>> query = "latest advancements in AI"
>>> result = retrieve_information_from_urls(urls, query)
"""
aggregated_content = ''
# Scrape and aggregate content from each URL
for url in urls:
scraped_content = Firecrawl().scrape(url)
aggregated_content += scraped_content["markdown"]
# Set up a vector retriever with local storage and embedding model from Mistral AI
auto_retriever = AutoRetriever(
vector_storage_local_path="local_data",
storage_type=StorageType.QDRANT,
embedding_model=MistralEmbedding(),
)
# Retrieve the most relevant information based on the query
# You can adjust the top_k and similarity_threshold value based on your needs
retrieved_info = auto_retriever.run_vector_retriever(
query=query,
contents=aggregated_content,
top_k=3,
similarity_threshold=0.5,
)
return retrieved_info
Letโs put the retrieval function to the test by gathering some information about the 2024 Olympics. The first run may take about 50 seconds as it needs to build a local vector database.
[15]:
retrieved_info = retrieve_information_from_urls(
query="Which country won the most golden prize in 2024 Olympics?",
urls=[
"https://www.nbcnews.com/sports/olympics/united-states-china-gold-medals-rcna166013",
],
)
print(retrieved_info)
{'Original Query': 'Which country won the most golden prize in 2024 Olympics?', 'Retrieved Context': [' most medals since 1992](https://olympics.com/en/olympic-games/barcelona-1992/medals), when the โUnified Team,โ\xa0athletes from the former Soviet Union republics, won 112 medals, including 45 golds. The Americans made 108 podium visits in Barcelona, with 37 of them on the top step.\n\nGold medalist Si', ' Olympic medals โ collected seven here.\n\nEcuador won five medals,more than doubling its overall summer total.\n\nKyrgyzstan made six podium trips, which is impressive given the central Asian nation came to Paris with only seven Olympic medals in history.\n\nBefore Paris 2024, Moldova had won six in national history. The tiny former Soviet republicmade four podium tripsin Paris.\n\n_David K. Li reported from Paris and Sean Nevin from London._\n\n[.\n\nWhile there had been no gold medal ties in Summer Games history, there have been three draws in the Winter Olympics.\n\nThe most recent draw was between Norway and Germany, each [winning 14\xa0golds in PyeongChang 2018](https://olympics.com/en/olympic-games/pyeongchang-2018/medals).\n\nThe U.S. still came out with the most overall medals, taking home 126 from Paris this year.\n\nThe U.S. hasnโt failed to win [the']}
๐ Thanks to CAMELโs RAG pipeline and Firecrawlโs tidy scraping capabilities, this function effectively retrieves relevant information from the specified URLs! You can now integrate this function into CAMELโs Agents to automate the retrieval process further.
๐น Monitoring AI Agents with AgentOps#
[16]:
import agentops
agentops.init(default_tags=["CAMEL cookbook"])
๐ AgentOps: Session Replay: https://app.agentops.ai/drilldown?session_id=ab0c011f-1928-4895-9747-ea882e9dfc38
[16]:
<agentops.session.Session at 0x7c82f2b71ae0>
๐ง Knowledge Graph Construction#
A powerful feature of CAMEL is its ability to build and store knowledge graphs from text data. This allows for advanced analysis and visualization of relationships within the data.
Set up your Neo4j instance by providing the URL, username, and password, here is the guidance, check your credentials in the downloaded .txt file. Note that you may need to wait up to 60 seconds if the instance has just been set up.
[27]:
from camel.storages import Neo4jGraph
from camel.loaders import UnstructuredIO
from camel.agents import KnowledgeGraphAgent
def knowledge_graph_builder(text_input: str) -> None:
r"""Build and store a knowledge graph from the provided text.
This function processes the input text to create and extract nodes and relationships,
which are then added to a Neo4j database as a knowledge graph.
Args:
text_input (str): The input text from which the knowledge graph is to be constructed.
Returns:
graph_elements: The generated graph element from knowlegde graph agent.
"""
# Set Neo4j instance
n4j = Neo4jGraph(
url="Your_URL",
username="Your_Username",
password="Your_Password",
)
# Initialize instances
uio = UnstructuredIO()
kg_agent = KnowledgeGraphAgent(model=mistral_large_2)
# Create an element from the provided text
element_example = uio.create_element_from_text(text=text_input, element_id="001")
# Extract nodes and relationships using the Knowledge Graph Agent
graph_elements = kg_agent.run(element_example, parse_graph_elements=True)
# Add the extracted graph elements to the Neo4j database
n4j.add_graph_elements(graph_elements=[graph_elements])
return graph_elements
๐ค๐ค Multi-Agent Role-Playing with CAMEL#
This section sets up a role-playing session where AI agents interact to accomplish a task using various tools. We will guide the assistant agent to perform a comprehensive study of the Turkish shooter in the 2024 Paris Olympics.
[28]:
from typing import List
from colorama import Fore
from camel.agents.chat_agent import FunctionCallingRecord
from camel.societies import RolePlaying
from camel.utils import print_text_animated
Defining the Task Prompt
[29]:
task_prompt = """Do a comprehensive study of the Turkish shooter in 2024 paris
olympics, write a report for me, then create a knowledge graph for the report.
You should use search tool to get related URLs first, then use retrieval tool
to get the retrieved content back by providing the list of URLs, finially
use tool to build the knowledge graph to finish the task.
No more other actions needed"""
We will configure the assistant agent with tools for mathematical calculations, web information retrieval, and knowledge graph building.
[30]:
retrieval_tool = OpenAIFunction(retrieve_information_from_urls)
search_tool = OpenAIFunction(SearchToolkit().search_duckduckgo)
knowledge_graph_tool = OpenAIFunction(knowledge_graph_builder)
tool_list = [
retrieval_tool,
search_tool,
knowledge_graph_tool,
]
assistant_model_config = MistralConfig(
tools=tool_list,
temperature=0.0,
)
Setting Up the Role-Playing Session
[31]:
# Initialize the role-playing session
role_play_session = RolePlaying(
assistant_role_name="CAMEL Assistant",
user_role_name="CAMEL User",
assistant_agent_kwargs=dict(
model=ModelFactory.create(
model_platform=ModelPlatformType.MISTRAL,
model_type=ModelType.MISTRAL_LARGE,
model_config_dict=assistant_model_config.as_dict(),
),
tools=tool_list,
),
user_agent_kwargs=dict(model=mistral_large_2),
task_prompt=task_prompt,
with_task_specify=False,
)
Print the system message and task prompt
[32]:
# Print system and task messages
print(
Fore.GREEN
+ f"AI Assistant sys message:\n{role_play_session.assistant_sys_msg}\n"
)
print(Fore.BLUE + f"AI User sys message:\n{role_play_session.user_sys_msg}\n")
print(Fore.YELLOW + f"Original task prompt:\n{task_prompt}\n")
print(
Fore.CYAN
+ "Specified task prompt:"
+ f"\n{role_play_session.specified_task_prompt}\n"
)
print(Fore.RED + f"Final task prompt:\n{role_play_session.task_prompt}\n")
AI Assistant sys message:
BaseMessage(role_name='CAMEL Assistant', role_type=<RoleType.ASSISTANT: 'assistant'>, meta_dict={'task': 'Do a comprehensive study of the Turkish shooter in 2024 paris\nolympics, write a report for me, then create a knowledge graph for the report.\nYou should use search tool to get related URLs first, then use retrieval tool\nto get the retrieved content back by providing the list of URLs, finially\nuse tool to build the knowledge graph to finish the task.\nNo more other actions needed', 'assistant_role': 'CAMEL Assistant', 'user_role': 'CAMEL User'}, content='===== RULES OF ASSISTANT =====\nNever forget you are a CAMEL Assistant and I am a CAMEL User. Never flip roles! Never instruct me!\nWe share a common interest in collaborating to successfully complete a task.\nYou must help me to complete the task.\nHere is the task: Do a comprehensive study of the Turkish shooter in 2024 paris\nolympics, write a report for me, then create a knowledge graph for the report.\nYou should use search tool to get related URLs first, then use retrieval tool\nto get the retrieved content back by providing the list of URLs, finially\nuse tool to build the knowledge graph to finish the task.\nNo more other actions needed. Never forget our task!\nI must instruct you based on your expertise and my needs to complete the task.\n\nI must give you one instruction at a time.\nYou must write a specific solution that appropriately solves the requested instruction and explain your solutions.\nYou must decline my instruction honestly if you cannot perform the instruction due to physical, moral, legal reasons or your capability and explain the reasons.\nUnless I say the task is completed, you should always start with:\n\nSolution: <YOUR_SOLUTION>\n\n<YOUR_SOLUTION> should be very specific, include detailed explanations and provide preferable detailed implementations and examples and lists for task-solving.\nAlways end <YOUR_SOLUTION> with: Next request.', video_bytes=None, image_list=None, image_detail='auto', video_detail='low')
AI User sys message:
BaseMessage(role_name='CAMEL User', role_type=<RoleType.USER: 'user'>, meta_dict={'task': 'Do a comprehensive study of the Turkish shooter in 2024 paris\nolympics, write a report for me, then create a knowledge graph for the report.\nYou should use search tool to get related URLs first, then use retrieval tool\nto get the retrieved content back by providing the list of URLs, finially\nuse tool to build the knowledge graph to finish the task.\nNo more other actions needed', 'assistant_role': 'CAMEL Assistant', 'user_role': 'CAMEL User'}, content='===== RULES OF USER =====\nNever forget you are a CAMEL User and I am a CAMEL Assistant. Never flip roles! You will always instruct me.\nWe share a common interest in collaborating to successfully complete a task.\nI must help you to complete the task.\nHere is the task: Do a comprehensive study of the Turkish shooter in 2024 paris\nolympics, write a report for me, then create a knowledge graph for the report.\nYou should use search tool to get related URLs first, then use retrieval tool\nto get the retrieved content back by providing the list of URLs, finially\nuse tool to build the knowledge graph to finish the task.\nNo more other actions needed. Never forget our task!\nYou must instruct me based on my expertise and your needs to solve the task ONLY in the following two ways:\n\n1. Instruct with a necessary input:\nInstruction: <YOUR_INSTRUCTION>\nInput: <YOUR_INPUT>\n\n2. Instruct without any input:\nInstruction: <YOUR_INSTRUCTION>\nInput: None\n\nThe "Instruction" describes a task or question. The paired "Input" provides further context or information for the requested "Instruction".\n\nYou must give me one instruction at a time.\nI must write a response that appropriately solves the requested instruction.\nI must decline your instruction honestly if I cannot perform the instruction due to physical, moral, legal reasons or my capability and explain the reasons.\nYou should instruct me not ask me questions.\nNow you must start to instruct me using the two ways described above.\nDo not add anything else other than your instruction and the optional corresponding input!\nKeep giving me instructions and necessary inputs until you think the task is completed.\nWhen the task is completed, you must only reply with a single word <CAMEL_TASK_DONE>.\nNever say <CAMEL_TASK_DONE> unless my responses have solved your task.', video_bytes=None, image_list=None, image_detail='auto', video_detail='low')
Original task prompt:
Do a comprehensive study of the Turkish shooter in 2024 paris
olympics, write a report for me, then create a knowledge graph for the report.
You should use search tool to get related URLs first, then use retrieval tool
to get the retrieved content back by providing the list of URLs, finially
use tool to build the knowledge graph to finish the task.
No more other actions needed
Specified task prompt:
None
Final task prompt:
Do a comprehensive study of the Turkish shooter in 2024 paris
olympics, write a report for me, then create a knowledge graph for the report.
You should use search tool to get related URLs first, then use retrieval tool
to get the retrieved content back by providing the list of URLs, finially
use tool to build the knowledge graph to finish the task.
No more other actions needed
Set the termination rule and start the interaction between agents
NOTE: This session will take approximately 8 minutes and will consume around 60k tokens by using Mistral Large 2 Model.
[33]:
n = 0
input_msg = role_play_session.init_chat()
while n < 20: # Limit the chat to 20 turns
n += 1
assistant_response, user_response = role_play_session.step(input_msg)
if assistant_response.terminated:
print(
Fore.GREEN
+ (
"AI Assistant terminated. Reason: "
f"{assistant_response.info['termination_reasons']}."
)
)
break
if user_response.terminated:
print(
Fore.GREEN
+ (
"AI User terminated. "
f"Reason: {user_response.info['termination_reasons']}."
)
)
break
# Print output from the user
print_text_animated(
Fore.BLUE + f"AI User:\n\n{user_response.msg.content}\n",
0.01
)
if "CAMEL_TASK_DONE" in user_response.msg.content:
break
# Print output from the assistant, including any function
# execution information
print_text_animated(Fore.GREEN + "AI Assistant:", 0.01)
tool_calls: List[FunctionCallingRecord] = [
FunctionCallingRecord(**call.as_dict())
for call in assistant_response.info['tool_calls']
]
for func_record in tool_calls:
print_text_animated(f"{func_record}", 0.01)
print_text_animated(f"{assistant_response.msg.content}\n", 0.01)
input_msg = assistant_response.msg
AI User:
Instruction: Use the search tool to find related URLs about the Turkish shooter in the 2024 Paris Olympics.
Input: None
AI Assistant:
Function Execution: search_duckduckgo
Args: {'query': 'Turkish shooter in the 2024 Paris Olympics'}
Result: [{'result_id': 1, 'title': 'Meet Yusuf Dikec: The Turkish shooter who has gone viral at Paris 2024 ...', 'description': "Yusuf Dikec, experienced Olympian and Turkish record breaker. Paris is far from Dikec's first rodeo. The Turkish shooter made his Olympic debut at Beijing 2008 aged 35, and returned four years later at London 2012.It was there where he recorded his previous best result of 13 th in the men's 50m free pistol event.. Dikec then competed at Rio 2016 and Tokyo 2020 held in 2021, though always shy ...", 'url': 'https://olympics.com/en/news/yusuf-dikec-turkish-shooting-viral-paris-2024-olympics'}, {'result_id': 2, 'title': 'Turkish sharpshooter Yusuf Dikeรง wins silver at Paris Olympics with ...', 'description': "Turkish sharpshooter Yusuf Dikeรง didn't need special lenses or big ear protectors to win the silver medal in the air pistol team event at the 2024 Paris Olympics.", 'url': 'https://www.nbcnews.com/news/sports/sharpshooter-yusuf-dikec-silver-olympics-minimal-gear-internet-sensati-rcna164685'}, {'result_id': 3, 'title': 'Who is Yusuf Dikec, Turkish shooter who went viral at 2024 Olympics ...', 'description': 'CHATEAUROUX, France (AP) โ Turkish pistol shooter Yusuf Dikec has gone viral on social media for his seemingly casual attitude while shooting his way to a silver medal at the 2024 Olympics. The most-shared images show Dikec shooting in a T-shirt with one hand in his pocket, a seemingly standard pair of glasses and an impassive look on his face.', 'url': 'https://apnews.com/article/olympics-2024-yusuf-dikec-turkish-shooter-a7890124304080a48e7ee4294004d306'}, {'result_id': 4, 'title': "WATCH: Turkey's sharpshooter wins silver without high-tech aids - NBC News", 'description': 'Turkish Olympian Yusuf Dikec won silver in the 2024 Paris Olympics mixed air pistol team event. Dikec went viral for his nonchalant approach of not using any high-tech aids in the competition.', 'url': 'https://www.nbcnews.com/video/watch-turkey-s-sharpshooter-wins-silver-without-high-tech-aids-216173637766'}, {'result_id': 5, 'title': "Yusuf Dikeรง: Turkey's viral Olympian inspires new celebration with ...", 'description': "Turkey's Yusuf Dikec competes in the shooting 10m air pistol mixed team gold medal match during the Paris 2024 Olympic Games at Chateauroux Shooting Centre on July 30, 2024. (Photo by Alain JOCARD ...", 'url': 'https://www.cnn.com/2024/08/09/sport/yusuf-dikec-olympic-celebration-craze-spt-intl/index.html'}]
Solution: I have used the search tool to find related URLs about the Turkish shooter in the 2024 Paris Olympics. Here are the results:
1. **Title:** Meet Yusuf Dikec: The Turkish shooter who has gone viral at Paris 2024 ...
**Description:** Yusuf Dikec, experienced Olympian and Turkish record breaker. Paris is far from Dikec's first rodeo. The Turkish shooter made his Olympic debut at Beijing 2008 aged 35, and returned four years later at London 2012. It was there where he recorded his previous best result of 13th in the men's 50m free pistol event. Dikec then competed at Rio 2016 and Tokyo 2020 held in 2021, though always shy ...
**URL:** [https://olympics.com/en/news/yusuf-dikec-turkish-shooting-viral-paris-2024-olympics](https://olympics.com/en/news/yusuf-dikec-turkish-shooting-viral-paris-2024-olympics)
2. **Title:** Turkish sharpshooter Yusuf Dikeรง wins silver at Paris Olympics with ...
**Description:** Turkish sharpshooter Yusuf Dikeรง didn't need special lenses or big ear protectors to win the silver medal in the air pistol team event at the 2024 Paris Olympics.
**URL:** [https://www.nbcnews.com/news/sports/sharpshooter-yusuf-dikec-silver-olympics-minimal-gear-internet-sensati-rcna164685](https://www.nbcnews.com/news/sports/sharpshooter-yusuf-dikec-silver-olympics-minimal-gear-internet-sensati-rcna164685)
3. **Title:** Who is Yusuf Dikec, Turkish shooter who went viral at 2024 Olympics ...
**Description:** CHATEAUROUX, France (AP) โ Turkish pistol shooter Yusuf Dikec has gone viral on social media for his seemingly casual attitude while shooting his way to a silver medal at the 2024 Olympics. The most-shared images show Dikec shooting in a T-shirt with one hand in his pocket, a seemingly standard pair of glasses and an impassive look on his face.
**URL:** [https://apnews.com/article/olympics-2024-yusuf-dikec-turkish-shooter-a7890124304080a48e7ee4294004d306](https://apnews.com/article/olympics-2024-yusuf-dikec-turkish-shooter-a7890124304080a48e7ee4294004d306)
4. **Title:** WATCH: Turkey's sharpshooter wins silver without high-tech aids - NBC News
**Description:** Turkish Olympian Yusuf Dikec won silver in the 2024 Paris Olympics mixed air pistol team event. Dikec went viral for his nonchalant approach of not using any high-tech aids in the competition.
**URL:** [https://www.nbcnews.com/video/watch-turkey-s-sharpshooter-wins-silver-without-high-tech-aids-216173637766](https://www.nbcnews.com/video/watch-turkey-s-sharpshooter-wins-silver-without-high-tech-aids-216173637766)
5. **Title:** Yusuf Dikeรง: Turkey's viral Olympian inspires new celebration with ...
**Description:** Turkey's Yusuf Dikec competes in the shooting 10m air pistol mixed team gold medal match during the Paris 2024 Olympic Games at Chateauroux Shooting Centre on July 30, 2024. (Photo by Alain JOCARD ...
**URL:** [https://www.cnn.com/2024/08/09/sport/yusuf-dikec-olympic-celebration-craze-spt-intl/index.html](https://www.cnn.com/2024/08/09/sport/yusuf-dikec-olympic-celebration-craze-spt-intl/index.html)
Next request.
AI User:
Instruction: Use the retrieval tool to get the content from the provided URLs.
Input: [https://olympics.com/en/news/yusuf-dikec-turkish-shooting-viral-paris-2024-olympics, https://www.nbcnews.com/news/sports/sharpshooter-yusuf-dikec-silver-olympics-minimal-gear-internet-sensati-rcna164685, https://apnews.com/article/olympics-2024-yusuf-dikec-turkish-shooter-a7890124304080a48e7ee4294004d306, https://www.nbcnews.com/video/watch-turkey-s-sharpshooter-wins-silver-without-high-tech-aids-216173637766, https://www.cnn.com/2024/08/09/sport/yusuf-dikec-olympic-celebration-craze-spt-intl/index.html]
AI Assistant:
Function Execution: retrieve_information_from_urls
Args: {'query': 'Turkish shooter in the 2024 Paris Olympics', 'urls': ['https://olympics.com/en/news/yusuf-dikec-turkish-shooting-viral-paris-2024-olympics', 'https://www.nbcnews.com/news/sports/sharpshooter-yusuf-dikec-silver-olympics-minimal-gear-internet-sensati-rcna164685', 'https://apnews.com/article/olympics-2024-yusuf-dikec-turkish-shooter-a7890124304080a48e7ee4294004d306', 'https://www.nbcnews.com/video/watch-turkey-s-sharpshooter-wins-silver-without-high-tech-aids-216173637766', 'https://www.cnn.com/2024/08/09/sport/yusuf-dikec-olympic-celebration-craze-spt-intl/index.html']}
Result: {'Original Query': 'Turkish shooter in the 2024 Paris Olympics', 'Retrieved Context': ["1/mobile/left.svg)\n\n\n\n\n\n## **Yusuf Dikec, experienced Olympian and Turkish record breaker**\n\nParis is far from Dikec's first rodeo. The Turkish shooter made his Olympic debut at [Beijing 2008](https://olympics.com/en/olympic-games/beijing-2008) aged 35, and returned four years later at [London 2012](https:", "to internet stardom.\n\nWithout the special lenses and ear protectors typical of the sport, and with one hand casually tucked into his pocket, Dikec effortlessly shot his way onto the podium. He and partner Sevval Ilayda Tarhan nabbed Turkeyโs first medal in Olympic shooting.\n\nTurkey's Yusuf Dikec competes in the shooting 10m air p", "ular pair of glasses. His partner, ลevval ฤฐlayda\xa0Tarhan, also didn't wear any blinders and kept one hand tucked in her pocket. Their win was Turkeyโs first medal in an Olympic shooting event.\n\nSilver medalists ลevval ฤฐlayda Tarhan and Yusuf Dikeรง of Turkey on the podium during after the 10-meter air pistol mixed "]}
Solution: I have used the retrieval tool to get the content from the provided URLs. Here is the retrieved content:
1. **URL:** [https://olympics.com/en/news/yusuf-dikec-turkish-shooting-viral-paris-2024-olympics](https://olympics.com/en/news/yusuf-dikec-turkish-shooting-viral-paris-2024-olympics)
**Content:**
```
## **Yusuf Dikec, experienced Olympian and Turkish record breaker**
Paris is far from Dikec's first rodeo. The Turkish shooter made his Olympic debut at [Beijing 2008](https://olympics.com/en/olympic-games/beijing-2008) aged 35, and returned four years later at [London 2012](https:
```
2. **URL:** [https://www.nbcnews.com/news/sports/sharpshooter-yusuf-dikec-silver-olympics-minimal-gear-internet-sensati-rcna164685](https://www.nbcnews.com/news/sports/sharpshooter-yusuf-dikec-silver-olympics-minimal-gear-internet-sensati-rcna164685)
**Content:**
```
Without the special lenses and ear protectors typical of the sport, and with one hand casually tucked into his pocket, Dikec effortlessly shot his way onto the podium. He and partner Sevval Ilayda Tarhan nabbed Turkeyโs first medal in Olympic shooting.
Turkey's Yusuf Dikec competes in the shooting 10m air pistol
```
3. **URL:** [https://apnews.com/article/olympics-2024-yusuf-dikec-turkish-shooter-a7890124304080a48e7ee4294004d306](https://apnews.com/article/olympics-2024-yusuf-dikec-turkish-shooter-a7890124304080a48e7ee4294004d306)
**Content:**
```
With a regular pair of glasses. His partner, ลevval ฤฐlayda Tarhan, also didnโt wear any blinders and kept one hand tucked in her pocket. Their win was Turkeyโs first medal in an Olympic shooting event.
Silver medalists ลevval ฤฐlayda Tarhan and Yusuf Dikeรง of Turkey on the podium during after the 10-meter air pistol mixed
```
Next request.
AI User:
Instruction: Summarize the key points from the retrieved content about the Turkish shooter Yusuf Dikec at the 2024 Paris Olympics.
Input: None
AI Assistant:
Solution: I have summarized the key points from the retrieved content about the Turkish shooter Yusuf Dikec at the 2024 Paris Olympics. Here are the main points:
1. **Experienced Olympian:** Yusuf Dikec is an experienced Olympian who made his debut at the Beijing 2008 Olympics at the age of 35. He has also competed in the London 2012, Rio 2016, and Tokyo 2020 Olympics.
2. **Silver Medal Win:** Dikec won a silver medal in the air pistol team event at the 2024 Paris Olympics. He competed with minimal gear, without special lenses or big ear protectors, and with one hand casually tucked into his pocket.
3. **Viral Sensation:** Dikec's nonchalant approach and minimal gear during the competition went viral on the internet. His seemingly casual attitude while shooting his way to a silver medal caught the attention of many.
4. **Partnership with Sevval Ilayda Tarhan:** Dikec competed alongside Sevval Ilayda Tarhan in the mixed air pistol team event. Together, they won Turkeyโs first medal in Olympic shooting.
5. **Historic Achievement:** The win marked Turkeyโs first medal in an Olympic shooting event, highlighting the significance of Dikec and Tarhan's achievement.
Next request.
AI User:
Instruction: Use the summarized key points to write a comprehensive report about the Turkish shooter Yusuf Dikec at the 2024 Paris Olympics.
Input: None
AI Assistant:
Solution: I have written a comprehensive report about the Turkish shooter Yusuf Dikec at the 2024 Paris Olympics based on the summarized key points. Here is the report:
---
### Comprehensive Report on Turkish Shooter Yusuf Dikec at the 2024 Paris Olympics
#### Introduction
Yusuf Dikec, a seasoned Olympian and Turkish record breaker, made headlines at the 2024 Paris Olympics with his impressive performance and unique style. This report delves into his background, achievements, and the impact he made during the games.
#### Background
Yusuf Dikec is an experienced shooter who has represented Turkey in multiple Olympic Games. He made his Olympic debut at the Beijing 2008 Olympics at the age of 35. Since then, he has competed in the London 2012, Rio 2016, and Tokyo 2020 Olympics, showcasing his skills and dedication to the sport.
#### Achievements at the 2024 Paris Olympics
##### Silver Medal Win
At the 2024 Paris Olympics, Yusuf Dikec won a silver medal in the air pistol team event. His performance was notable not only for his accuracy but also for his minimalistic approach. Unlike many of his competitors, Dikec did not use special lenses or big ear protectors. Instead, he competed with a regular pair of glasses and one hand casually tucked into his pocket, demonstrating a level of confidence and skill that captivated audiences.
##### Viral Sensation
Dikec's nonchalant attitude and minimal gear during the competition went viral on the internet. Images and videos of him shooting effortlessly while maintaining a seemingly casual demeanor spread rapidly across social media platforms. His unique style and impressive performance made him an internet sensation, drawing attention from both sports enthusiasts and casual viewers alike.
##### Partnership with Sevval Ilayda Tarhan
Dikec competed alongside Sevval Ilayda Tarhan in the mixed air pistol team event. Together, they secured Turkeyโs first medal in Olympic shooting. Tarhan, like Dikec, also competed with minimal gear, keeping one hand tucked in her pocket and not wearing any blinders. Their partnership and shared approach to the sport highlighted their synergy and skill.
#### Historic Achievement
The silver medal win marked a historic achievement for Turkey. It was the countryโs first medal in an Olympic shooting event, underscoring the significance of Dikec and Tarhan's performance. Their success not only brought pride to their nation but also inspired a new generation of shooters in Turkey.
#### Conclusion
Yusuf Dikec's journey at the 2024 Paris Olympics is a testament to his skill, dedication, and unique style. His silver medal win, viral fame, and historic achievement have cemented his place in the annals of Turkish sports history. As he continues to compete, Dikec serves as an inspiration for aspiring athletes, showcasing the power of perseverance and individuality in sports.
---
Next request.
AI User:
Instruction: Use the tool to build a knowledge graph based on the comprehensive report about the Turkish shooter Yusuf Dikec at the 2024 Paris Olympics.
Input: None
AI Assistant:
Function Execution: knowledge_graph_builder
Args: {'text_input': "### Comprehensive Report on Turkish Shooter Yusuf Dikec at the 2024 Paris Olympics\n\n#### Introduction\n\nYusuf Dikec, a seasoned Olympian and Turkish record breaker, made headlines at the 2024 Paris Olympics with his impressive performance and unique style. This report delves into his background, achievements, and the impact he made during the games.\n\n#### Background\n\nYusuf Dikec is an experienced shooter who has represented Turkey in multiple Olympic Games. He made his Olympic debut at the Beijing 2008 Olympics at the age of 35. Since then, he has competed in the London 2012, Rio 2016, and Tokyo 2020 Olympics, showcasing his skills and dedication to the sport.\n\n#### Achievements at the 2024 Paris Olympics\n\n##### Silver Medal Win\n\nAt the 2024 Paris Olympics, Yusuf Dikec won a silver medal in the air pistol team event. His performance was notable not only for his accuracy but also for his minimalistic approach. Unlike many of his competitors, Dikec did not use special lenses or big ear protectors. Instead, he competed with a regular pair of glasses and one hand casually tucked into his pocket, demonstrating a level of confidence and skill that captivated audiences.\n\n##### Viral Sensation\n\nDikec's nonchalant attitude and minimal gear during the competition went viral on the internet. Images and videos of him shooting effortlessly while maintaining a seemingly casual demeanor spread rapidly across social media platforms. His unique style and impressive performance made him an internet sensation, drawing attention from both sports enthusiasts and casual viewers alike.\n\n##### Partnership with Sevval Ilayda Tarhan\n\nDikec competed alongside Sevval Ilayda Tarhan in the mixed air pistol team event. Together, they secured Turkeyโs first medal in Olympic shooting. Tarhan, like Dikec, also competed with minimal gear, keeping one hand tucked in her pocket and not wearing any blinders. Their partnership and shared approach to the sport highlighted their synergy and skill.\n\n#### Historic Achievement\n\nThe silver medal win marked a historic achievement for Turkey. It was the countryโs first medal in an Olympic shooting event, underscoring the significance of Dikec and Tarhan's performance. Their success not only brought pride to their nation but also inspired a new generation of shooters in Turkey.\n\n#### Conclusion\n\nYusuf Dikec's journey at the 2024 Paris Olympics is a testament to his skill, dedication, and unique style. His silver medal win, viral fame, and historic achievement have cemented his place in the annals of Turkish sports history. As he continues to compete, Dikec serves as an inspiration for aspiring athletes, showcasing the power of perseverance and individuality in sports."}
Result: {'nodes': [{'id': 'Yusuf Dikec', 'type': 'Person', 'properties': {'source': 'agent_created'}}, {'id': 'Turkey', 'type': 'Country', 'properties': {'source': 'agent_created'}}, {'id': '2024 Paris Olympics', 'type': 'Event', 'properties': {'source': 'agent_created'}}, {'id': 'Silver Medal', 'type': 'Achievement', 'properties': {'source': 'agent_created'}}, {'id': 'Air Pistol Team Event', 'type': 'Event', 'properties': {'source': 'agent_created'}}, {'id': 'Sevval Ilayda Tarhan', 'type': 'Person', 'properties': {'source': 'agent_created'}}, {'id': 'Mixed Air Pistol Team Event', 'type': 'Event', 'properties': {'source': 'agent_created'}}, {'id': 'Beijing 2008 Olympics', 'type': 'Event', 'properties': {'source': 'agent_created'}}, {'id': 'London 2012 Olympics', 'type': 'Event', 'properties': {'source': 'agent_created'}}, {'id': 'Rio 2016 Olympics', 'type': 'Event', 'properties': {'source': 'agent_created'}}, {'id': 'Tokyo 2020 Olympics', 'type': 'Event', 'properties': {'source': 'agent_created'}}], 'relationships': [{'subj': {'id': 'Yusuf Dikec', 'type': 'Person', 'properties': {'source': 'agent_created'}}, 'obj': {'id': 'Turkey', 'type': 'Country', 'properties': {'source': 'agent_created'}}, 'type': 'Represents', 'properties': {'source': 'agent_created'}}, {'subj': {'id': 'Yusuf Dikec', 'type': 'Person', 'properties': {'source': 'agent_created'}}, 'obj': {'id': '2024 Paris Olympics', 'type': 'Event', 'properties': {'source': 'agent_created'}}, 'type': 'ParticipatedIn', 'properties': {'source': 'agent_created'}}, {'subj': {'id': 'Yusuf Dikec', 'type': 'Person', 'properties': {'source': 'agent_created'}}, 'obj': {'id': 'Silver Medal', 'type': 'Achievement', 'properties': {'source': 'agent_created'}}, 'type': 'Won', 'properties': {'source': 'agent_created'}}, {'subj': {'id': 'Yusuf Dikec', 'type': 'Person', 'properties': {'source': 'agent_created'}}, 'obj': {'id': 'Air Pistol Team Event', 'type': 'Event', 'properties': {'source': 'agent_created'}}, 'type': 'CompetedIn', 'properties': {'source': 'agent_created'}}, {'subj': {'id': 'Yusuf Dikec', 'type': 'Person', 'properties': {'source': 'agent_created'}}, 'obj': {'id': 'Sevval Ilayda Tarhan', 'type': 'Person', 'properties': {'source': 'agent_created'}}, 'type': 'PartneredWith', 'properties': {'source': 'agent_created'}}, {'subj': {'id': 'Yusuf Dikec', 'type': 'Person', 'properties': {'source': 'agent_created'}}, 'obj': {'id': 'Mixed Air Pistol Team Event', 'type': 'Event', 'properties': {'source': 'agent_created'}}, 'type': 'CompetedIn', 'properties': {'source': 'agent_created'}}, {'subj': {'id': 'Yusuf Dikec', 'type': 'Person', 'properties': {'source': 'agent_created'}}, 'obj': {'id': 'Beijing 2008 Olympics', 'type': 'Event', 'properties': {'source': 'agent_created'}}, 'type': 'ParticipatedIn', 'properties': {'source': 'agent_created'}}, {'subj': {'id': 'Yusuf Dikec', 'type': 'Person', 'properties': {'source': 'agent_created'}}, 'obj': {'id': 'London 2012 Olympics', 'type': 'Event', 'properties': {'source': 'agent_created'}}, 'type': 'ParticipatedIn', 'properties': {'source': 'agent_created'}}, {'subj': {'id': 'Yusuf Dikec', 'type': 'Person', 'properties': {'source': 'agent_created'}}, 'obj': {'id': 'Rio 2016 Olympics', 'type': 'Event', 'properties': {'source': 'agent_created'}}, 'type': 'ParticipatedIn', 'properties': {'source': 'agent_created'}}, {'subj': {'id': 'Yusuf Dikec', 'type': 'Person', 'properties': {'source': 'agent_created'}}, 'obj': {'id': 'Tokyo 2020 Olympics', 'type': 'Event', 'properties': {'source': 'agent_created'}}, 'type': 'ParticipatedIn', 'properties': {'source': 'agent_created'}}, {'subj': {'id': 'Sevval Ilayda Tarhan', 'type': 'Person', 'properties': {'source': 'agent_created'}}, 'obj': {'id': 'Turkey', 'type': 'Country', 'properties': {'source': 'agent_created'}}, 'type': 'Represents', 'properties': {'source': 'agent_created'}}, {'subj': {'id': 'Sevval Ilayda Tarhan', 'type': 'Person', 'properties': {'source': 'agent_created'}}, 'obj': {'id': '2024 Paris Olympics', 'type': 'Event', 'properties': {'source': 'agent_created'}}, 'type': 'ParticipatedIn', 'properties': {'source': 'agent_created'}}, {'subj': {'id': 'Sevval Ilayda Tarhan', 'type': 'Person', 'properties': {'source': 'agent_created'}}, 'obj': {'id': 'Mixed Air Pistol Team Event', 'type': 'Event', 'properties': {'source': 'agent_created'}}, 'type': 'CompetedIn', 'properties': {'source': 'agent_created'}}, {'subj': {'id': 'Turkey', 'type': 'Country', 'properties': {'source': 'agent_created'}}, 'obj': {'id': 'Silver Medal', 'type': 'Achievement', 'properties': {'source': 'agent_created'}}, 'type': 'Achieved', 'properties': {'source': 'agent_created'}}, {'subj': {'id': 'Yusuf Dikec', 'type': 'Person', 'properties': {'source': 'agent_created'}}, 'obj': {'id': 'Turkey', 'type': 'Country', 'properties': {'source': 'agent_created'}}, 'type': 'Represents', 'properties': {'source': 'agent_created'}}, {'subj': {'id': 'Yusuf Dikec', 'type': 'Person', 'properties': {'source': 'agent_created'}}, 'obj': {'id': '2024 Paris Olympics', 'type': 'Event', 'properties': {'source': 'agent_created'}}, 'type': 'ParticipatedIn', 'properties': {'source': 'agent_created'}}, {'subj': {'id': 'Yusuf Dikec', 'type': 'Person', 'properties': {'source': 'agent_created'}}, 'obj': {'id': 'Silver Medal', 'type': 'Achievement', 'properties': {'source': 'agent_created'}}, 'type': 'Won', 'properties': {'source': 'agent_created'}}, {'subj': {'id': 'Yusuf Dikec', 'type': 'Person', 'properties': {'source': 'agent_created'}}, 'obj': {'id': 'Air Pistol Team Event', 'type': 'Event', 'properties': {'source': 'agent_created'}}, 'type': 'CompetedIn', 'properties': {'source': 'agent_created'}}, {'subj': {'id': 'Yusuf Dikec', 'type': 'Person', 'properties': {'source': 'agent_created'}}, 'obj': {'id': 'Sevval Ilayda Tarhan', 'type': 'Person', 'properties': {'source': 'agent_created'}}, 'type': 'PartneredWith', 'properties': {'source': 'agent_created'}}, {'subj': {'id': 'Yusuf Dikec', 'type': 'Person', 'properties': {'source': 'agent_created'}}, 'obj': {'id': 'Mixed Air Pistol Team Event', 'type': 'Event', 'properties': {'source': 'agent_created'}}, 'type': 'CompetedIn', 'properties': {'source': 'agent_created'}}, {'subj': {'id': 'Yusuf Dikec', 'type': 'Person', 'properties': {'source': 'agent_created'}}, 'obj': {'id': 'Beijing 2008 Olympics', 'type': 'Event', 'properties': {'source': 'agent_created'}}, 'type': 'ParticipatedIn', 'properties': {'source': 'agent_created'}}, {'subj': {'id': 'Yusuf Dikec', 'type': 'Person', 'properties': {'source': 'agent_created'}}, 'obj': {'id': 'London 2012 Olympics', 'type': 'Event', 'properties': {'source': 'agent_created'}}, 'type': 'ParticipatedIn', 'properties': {'source': 'agent_created'}}, {'subj': {'id': 'Yusuf Dikec', 'type': 'Person', 'properties': {'source': 'agent_created'}}, 'obj': {'id': 'Rio 2016 Olympics', 'type': 'Event', 'properties': {'source': 'agent_created'}}, 'type': 'ParticipatedIn', 'properties': {'source': 'agent_created'}}, {'subj': {'id': 'Yusuf Dikec', 'type': 'Person', 'properties': {'source': 'agent_created'}}, 'obj': {'id': 'Tokyo 2020 Olympics', 'type': 'Event', 'properties': {'source': 'agent_created'}}, 'type': 'ParticipatedIn', 'properties': {'source': 'agent_created'}}, {'subj': {'id': 'Sevval Ilayda Tarhan', 'type': 'Person', 'properties': {'source': 'agent_created'}}, 'obj': {'id': 'Turkey', 'type': 'Country', 'properties': {'source': 'agent_created'}}, 'type': 'Represents', 'properties': {'source': 'agent_created'}}, {'subj': {'id': 'Sevval Ilayda Tarhan', 'type': 'Person', 'properties': {'source': 'agent_created'}}, 'obj': {'id': '2024 Paris Olympics', 'type': 'Event', 'properties': {'source': 'agent_created'}}, 'type': 'ParticipatedIn', 'properties': {'source': 'agent_created'}}, {'subj': {'id': 'Sevval Ilayda Tarhan', 'type': 'Person', 'properties': {'source': 'agent_created'}}, 'obj': {'id': 'Mixed Air Pistol Team Event', 'type': 'Event', 'properties': {'source': 'agent_created'}}, 'type': 'CompetedIn', 'properties': {'source': 'agent_created'}}, {'subj': {'id': 'Turkey', 'type': 'Country', 'properties': {'source': 'agent_created'}}, 'obj': {'id': 'Silver Medal', 'type': 'Achievement', 'properties': {'source': 'agent_created'}}, 'type': 'Achieved', 'properties': {'source': 'agent_created'}}], 'source': <unstructured.documents.elements.Text object at 0x7c82e2847520>}
Solution: I have used the tool to build a knowledge graph based on the comprehensive report about the Turkish shooter Yusuf Dikec at the 2024 Paris Olympics. Here is the knowledge graph:
### Knowledge Graph
#### Nodes
1. **Yusuf Dikec** (Person)
2. **Turkey** (Country)
3. **2024 Paris Olympics** (Event)
4. **Silver Medal** (Achievement)
5. **Air Pistol Team Event** (Event)
6. **Sevval Ilayda Tarhan** (Person)
7. **Mixed Air Pistol Team Event** (Event)
8. **Beijing 2008 Olympics** (Event)
9. **London 2012 Olympics** (Event)
10. **Rio 2016 Olympics** (Event)
11. **Tokyo 2020 Olympics** (Event)
#### Relationships
1. **Yusuf Dikec** (Person) - **Represents** - **Turkey** (Country)
2. **Yusuf Dikec** (Person) - **ParticipatedIn** - **2024 Paris Olympics** (Event)
3. **Yusuf Dikec** (Person) - **Won** - **Silver Medal** (Achievement)
4. **Yusuf Dikec** (Person) - **CompetedIn** - **Air Pistol Team Event** (Event)
5. **Yusuf Dikec** (Person) - **PartneredWith** - **Sevval Ilayda Tarhan** (Person)
6. **Yusuf Dikec** (Person) - **CompetedIn** - **Mixed Air Pistol Team Event** (Event)
7. **Yusuf Dikec** (Person) - **ParticipatedIn** - **Beijing 2008 Olympics** (Event)
8. **Yusuf Dikec** (Person) - **ParticipatedIn** - **London 2012 Olympics** (Event)
9. **Yusuf Dikec** (Person) - **ParticipatedIn** - **Rio 2016 Olympics** (Event)
10. **Yusuf Dikec** (Person) - **ParticipatedIn** - **Tokyo 2020 Olympics** (Event)
11. **Sevval Ilayda Tarhan** (Person) - **Represents** - **Turkey** (Country)
12. **Sevval Ilayda Tarhan** (Person) - **ParticipatedIn** - **2024 Paris Olympics** (Event)
13. **Sevval Ilayda Tarhan** (Person) - **CompetedIn** - **Mixed Air Pistol Team Event** (Event)
14. **Turkey** (Country) - **Achieved** - **Silver Medal** (Achievement)
This knowledge graph captures the key entities and their relationships as described in the comprehensive report about Yusuf Dikec at the 2024 Paris Olympics.
Next request.
AI User:
<CAMEL_TASK_DONE>
[34]:
# End the AgentOps session
agentops.end_session("Success")
๐ AgentOps: This run's cost $0.00
๐ AgentOps: Session Replay: https://app.agentops.ai/drilldown?session_id=ab0c011f-1928-4895-9747-ea882e9dfc38
๐ Go to the AgentOps link shown above, you will be able to see the detailed record for this running like below.
NOTE: The AgentOps link is private and tied to the AgentOps account. To access the link, youโll need to run the session using your own AgentOps API Key, which will then allow you to open the link with the sessionโs running information.
Currently AgentOps canโt get the running cost for Mistral AI directly.
๐ You can also go the the Neo4j Aura to check the knowledge graph generated by CAMELโs Agent like below.
๐ Highlights#
This notebook has guided you through setting up and running a CAMEL RAG workflow with Firecrawl for a complex, multi-agent role-playing task. You can adapt and expand this example for various other scenarios requiring advanced web information retrieval and AI collaboration.
Key tools utilized in this notebook include:
CAMEL: A powerful multi-agent framework that enables Retrieval-Augmented Generation and multi-agent role-playing scenarios, allowing for sophisticated AI-driven tasks.
Mistral: Utilized for its state-of-the-art language models, which enable tool-calling capabilities to execute external functions, while its powerful embeddings are employed for semantic search and content retrieval.
Firecrawl: A robust web scraping tool that simplifies extracting and cleaning content from various web pages.
AgentOps: Track and analysis the running of CAMEL Agents.
Qdrant: An efficient vector storage system used with CAMELโs AutoRetriever to store and retrieve relevant information based on vector similarities.
Neo4j: A leading graph database management system used for constructing and storing knowledge graphs, enabling complex relationships between entities to be mapped and queried efficiently.
DuckDuckGo Search: Utilized within the SearchToolkit to gather relevant URLs and information from the web, serving as the primary search engine for retrieving initial content.
Unstructured IO: Used for content chunking, facilitating the management of unstructured data for more efficient processing.
This comprehensive setup allows you to adapt and expand the example for various scenarios requiring advanced web information retrieval, AI collaboration, and multi-source data aggregation.
CAMEL also support advanced GraphRAG, for more information please checkhere
โญ Star the Repo
If you find CAMEL useful or interesting, please consider giving it a star on GitHub! Your stars help others find this project and motivate us to continue improving it.