Create, Evaluate, and Connect AI Skills
Python ai-agentsai-skillsartificial-intelligenceknowledge-graphlarge-language-modelsnatural-language-processingopenkgskillnetskills
"""
Skill Creation Example - Using SkillNetClient
"""
import os
from skillnet_ai import SkillNetClient
# Sample trajectory
SAMPLE_TRAJECTORY = """
User: Please analyze this invoice.pdf and extract the total amount and date.
Agent: I will use the PDF tool. Opening invoice.pdf...
Agent: Text extracted. Found 'Total: $500' and 'Date: 2023-10-12'.
Agent: The total is $500 and the date is Oct 12, 2023.
"""
# Sample GitHub repository URL
SAMPLE_GITHUB_URL = "https://github.com/zjunlp/DeepKE"
def main():
# Initialize client
client = SkillNetClient(
api_key=os.getenv("API_KEY"),
base_url=os.getenv("BASE_URL", "https://api.openai.com/v1"),
github_token=os.getenv("GITHUB_TOKEN"),
)
# Example 1: Create from trajectory
print("π Creating skill from trajectory...")
paths = client.create(
trajectory_content=SAMPLE_TRAJECTORY,
output_dir="./created_skills"
)
for p in paths:
print(f" - {p}")
# Example 2: Create from GitHub repository
print("\nπ Creating skill from GitHub...")
paths = client.create(
github_url=SAMPLE_GITHUB_URL,
output_dir="./created_skills"
)
for p in paths:
print(f" - {p}")
if __name__ == "__main__":
main()