CrewAI FileReadTool() - error

research_eoi.py

import toml
import pandas as pd
import textwrap
import yaml
from crewai import Agent, Task, Crew, Process, LLM
from crewai.knowledge.source.string_knowledge_source import StringKnowledgeSource
from crewai.knowledge.source.pdf_knowledge_source import PDFKnowledgeSource

# Create a knowledge source
from crewai_tools import PDFSearchTool, FileReadTool


nihr_guidelines_rag = PDFSearchTool(pdf="knowledge/eoi_guidelines_nihr.pdf")
nihr_guidelines_content = FileReadTool(file_path="knowledge/eoi_guidelines_nihr.txt")

site_profile_rag = PDFSearchTool(pdf="knowledge/thechelseapractice.pdf")
site_profile_content = FileReadTool(file_path="knowledge/thechelseapractice.txt")

site_identification_questions = FileReadTool(file_path="knowledge/site_identification_questions.txt")

# 🅾️ Update study.txt with each new study or make a tool to extract the pdf version of the study docuyment.
study_rag = PDFSearchTool(pdf="knowledge/study.pdf")
study_content = FileReadTool(file_path="knowledge/study.txt")


# Create an LLM with a temperature of 0 to ensure deterministic outputs
llm = LLM(model="gpt-4o-mini", temperature=0)

with open("notioncrew/config_secrets.toml", "r") as f:
    config_secrets = toml.load(f)

# Load environment variables from streamlit secrets
OPENAI_API_KEY = config_secrets["OPENAI_API_KEY"]
OPENAI_MODEL_NAME = config_secrets["OPENAI_MODEL_NAME"]
NOTION_ENDPOINT = config_secrets["NOTION_ENDPOINT"]
NOTION_VERSION = config_secrets["NOTION_VERSION"]
NOTION_TOKEN = config_secrets["NOTION_TOKEN"]
NOTION_DATABASE_ID = config_secrets["NOTION_DATABASE_ID"]
SERPER_API_KEY = config_secrets["SERPER_API_KEY"]
APPRAISAL_DATABASE_ID = config_secrets["APPRAISAL_DATABASE_ID"]


# Define file paths for YAML configurations
files = {
    "agents": "notioncrew/config/agents.yaml",
    "tasks": "notioncrew/config/tasks.yaml",
}

# Load configurations from YAML files
configs = {}
for config_type, file_path in files.items():
    with open(file_path, "r") as file:
        configs[config_type] = yaml.safe_load(file)

# Assign loaded configurations to specific variables
agents_config = configs["agents"]
tasks_config = configs["tasks"]

clinical_researcher_agent = Agent(
    config=agents_config["clinical_researcher_agent"],
    tools=[study_rag],
)

research_coordinator_agent = Agent(
    config=agents_config["research_coordinator_agent"],
    tools=[site_profile_content, site_profile_rag, nihr_guidelines_content],
)

senior_research_writer_agent = Agent(
    config=agents_config["senior_research_writer_agent"],
    tools=[nihr_guidelines_content, site_identification_questions],
)

research_auditor_agent = Agent(
    config=agents_config["research_auditor_agent"],
    tools=[nihr_guidelines_content, nihr_guidelines_rag],
)

# TASKS
review_study_specifications = Task(
    config=tasks_config["review_study_specifications"],
    agent=clinical_researcher_agent,
    output_file="01_the_study.md",
)

review_research_site = Task(
    config=tasks_config["review_research_site"],
    agent=research_coordinator_agent,
    output_file="02_site_profile.md",
)

complete_site_identification_questionnaire = Task(
    config=tasks_config["complete_site_identification_questionnaire"],
    agent=senior_research_writer_agent,
    tools = [site_identification_questions],
    output_file="03_site_identification_questions.md",
)

audit_site_identification_questionnaire = Task(
    config=tasks_config["audit_site_identification_questionnaire"],
    agent=research_auditor_agent,
    output_file="04_submission_audit.md",
)

# Creating Crew
crew = Crew(
    agents=[
        clinical_researcher_agent,
        research_coordinator_agent,
        senior_research_writer_agent,
        research_auditor_agent,
    ],
    tasks=[
        review_study_specifications,
        review_research_site,
        complete_site_identification_questionnaire,
        audit_site_identification_questionnaire,
    ],
    process=Process.sequential,
    verbose=True,
)
inputs = {"study_identifier": "CN012-0023"}
result = crew.kickoff()


costs = (
    0.150
    * (crew.usage_metrics.prompt_tokens + crew.usage_metrics.completion_tokens)
    / 1_000_000
)
print(f"💷 **Total costs**: ${costs:.4f}")

# Convert UsageMetrics instance to a DataFrame
df_usage_metrics = pd.DataFrame([crew.usage_metrics.dict()])
print(df_usage_metrics)


markdown_text = result.raw