relevanceai 10.2.2

No project description provided

Relevance AI - The home of your AI Workforce

🔥 Use Relevance to build AI agents for your AI workforce

Sign up for a free account ->

🧠 Documentation

Type	Link
Home Page	Home Page
Platform	Platform
Developer Documentation	Documentation

Relevance AI SDK

Welcome to the Relevance AI SDK! This guide will help you set up and start using the SDK to interact with your AI agents, tools, and knowledge.

Installation

To get started, you'll need to install the RelevanceAI library in a Python 3 environment. Run the following command in your terminal:

pip install relevanceai

Create an Account

Before using the SDK, ensure you have an account with Relevance AI.

1. Sign up for a free account at Relevance AI and log in.
2. Create a new secret key at SDK Login. Scroll to the bottom of the integrations page, click on "+ Create new secret key," and select "Admin" permissions.

Set Up Your Client

To interact with Relevance AI, you'll need to set up a client. Start by importing the library:

from relevanceai import RelevanceAI
client = RelevanceAI()

Validate Client Credentials

You can validate your client credentials by storing them as environment variables and loading them into your project using python-dotenv or the os library:

RAI_API_KEY=
RAI_REGION=
RAI_PROJECT=

from dotenv import load_dotenv
load_dotenv()

from relevanceai import RelevanceAI
client = RelevanceAI()

Alternatively, pass the credentials directly to the client:

from relevanceai import RelevanceAI
client = RelevanceAI(
    api_key="your_api_key", 
    region="your_region", 
    project="your_project"
)

You are now ready to start using Relevance AI via the Python SDK.

Quickstart

Using Agents & Tasks

List all the agents in your project:

from relevanceai import RelevanceAI
client = RelevanceAI()

agents = client.agents.list_agents()
print(agents)
# Example output: [Agent(agent_id="xxxxxxxx", name="Sales Qualifier"), ...]

Retrieve and interact with a specific agent:

my_agent = client.agents.retrieve_agent(agent_id="xxxxxxxx")

message = "Let's qualify this lead:\n\nName: Ethan Trang\n\nCompany: Relevance AI\n\nEmail: [e-mail]"

# Trigger a task with the agent
task = my_agent.trigger_task(message=message)
print(f"Task started with ID: {task.conversation_id}")

# View task progress
steps = my_agent.view_task_steps(conversation_id=task.conversation_id)

Using Tools

List all the tools in your project:

tools = client.tools.list_tools()
print(tools)
# Example output: [Tool(tool_id="xxxxxxxx", title="Search Website"), ...]

Retrieve and interact with a specific tool:

my_tool = client.tools.retrieve_tool(tool_id="xxxxxxxx")

# Check tool parameters schema
params_schema = my_tool.get_params_schema()

# Trigger the tool
result = my_tool.trigger(params={"search_query": "AI automation"})

Managing Knowledge Sets

Work with knowledge sets to store and retrieve data:

# List knowledge sets
knowledge_sets = client.knowledge.list_knowledge()

# Retrieve data from a knowledge set
data = client.knowledge.retrieve_knowledge(knowledge_set="my_dataset")

Managing Tasks

Track and manage ongoing tasks:

# Get task metadata
metadata = client.tasks.get_metadata(conversation_id="xxxxxxxx")

# Delete a completed task
client.tasks.delete_task(conversation_id="xxxxxxxx")

Explore More

Explore all the methods available for agents, tasks, tools, and knowledge with the documentation