Agents
This page assumes you are already familiar with how get hold of an api_key
and org_id
and how to install the Engage Smarter Python library. If not, turn to the Getting Started tutorial.
The Engage Smarter AI Platform makes different AI financial guidance experts - or Agents - available via its API. Any agent can be selected and used with either the Conversations
or Runs
API endpoints.
Here we explore the Agents
API, discussing:
- How to view which agents are currently available
- How agents are versioned in the API
Pre-requisites
First, we initialise the Engage Smarter Python Client
.
import os
from engagesmarter import Client
api_key = os.getenv("ENGAGE_SMARTER_API_KEY")
org_id = os.getenv("ENGAGE_SMARTER_ORG_ID")
client = Client(
api_key=api_key,
org_id=org_id,
)
Viewing available agents
Once the Client
is initialised, we can use it to make API requests. To list the agents currently available, use the list
method on the Agents (agents
) API resource, as follows:
agents = client.agents.list()
Each entry contains:
- the
name
of the agent, which is the parameter to use for thename
argument in both conversations and runs. - the
title
of the agent, which is a more human-readable version of the name. - the
description
of the agent, which is a short description of the agent's capabilities. - a
welcome_message
, which is an optional greeting to use when integrating the agent into larger systems.
For instance, we can see the first agent in the list returned below.
print(agents[0])
Agent versioning
All agents are versioned using a date-based versioning scheme. For a given type of agent - for instance, the pensions-questions-*
agent - there are at least two versions available:
- Latest version. This uses the suffix
latest
to signal that it is the most up-to-date version of the agent. - Date-stamped version. This is the agent as of a particular date. The date is encoded in the suffix, using the format
YYMMDD
. For instance,pensions-questions-240214
would be the agent as of 14 February 2024.