Define the service

Given an existing agent blueprint, either developed by yourself on the previous step and stored in the local registry, or an agent blueprint fetched from the remote registry, the next step consists in defining AI agent and tuning its configuration options.

Part of the development process covered in this guide

What you will learn

This guide covers step 4 of the development process. You will learn how to define the AI agent itself. This consists in indicating what agent blueprint will the AI agent be using, and configuring the AI agent parameters through ai-agent-level overrides.

You must ensure that your machine satisfies the framework requirements, you have set up the framework, and you have a local registry populated with some default components. As a result you should have a Pipenv workspace folder with an initialized local registry (./packages) in it.

Step-by-step instructions

The process to define an AI agent configuration is similar if you are using an agent blueprint being developed by you (for example, the agent blueprint created in the previous guide), or an existing, third-party agent blueprint downloaded from the remote registry. In the example below, we will be using the hello_world agent blueprint from the remote registry, which can be found in the list of packages shipped with the framework, but you can replace it with your own agent blueprint.

Ensure that the agent blueprint required by your AI agent is in the local registry. Your agent blueprint, all its required components and their dependencies must be downloaded to the local registry. You can read how to add missing components to the local registry. If you have populated the local registry with the required components to follow these guides, you do not need to take any further action.

Create the AI agent configuration file. Create a folder for your AI agent in the local registry (./packages). Pay attention to the correct format of the folder:

mkdir -p ./packages/your_name/services/your_ai_agent/

Within the AI agent folder, create the AI agent configuration file service.yaml:

touch ./packages/your_name/services/your_ai_agent/service.yaml

This file must contain:

A number of mandatory parameters.

Important

Ensure that author and AI agent name match the path within the local registry.
A reference to the agent blueprint that the AI agent will be using.
Configuration overrides that specify values for component parameters. These overrides are separated by YAML document separators --- and will be discussed in a further section.

Example of a service.yaml file

This is a complete example of an AI agent configuration file that uses the hello_world agent blueprint and overrides some required component parameters.

service.yaml

name: your_ai_agent
author: your_name
version: 0.1.0
description: A demonstration of a simple AI agent
aea_version: '>=2.0.0, <3.0.0'
license: Apache-2.0
fingerprint: {}
fingerprint_ignore_patterns: []
agent: valory/hello_world:0.1.0
number_of_agents: 4
deployment: {}
---
public_id: valory/hello_world_abci:0.1.0
type: skill
models:
  params:
    args:
      service_registry_address: null
      share_tm_config_on_startup: false
      on_chain_service_id: null
      setup:
        all_participants: ${ALL_PARTICIPANTS:list:["0x0000000000000000000000000000000000000000"]}
        safe_contract_address: '0x0000000000000000000000000000000000000000'
        consensus_threshold: null
      hello_world_message: ${HELLO_WORLD_STRING_0:str:HELLO_WORLD!}
  benchmark_tool:
    args:
      log_dir: /benchmarks
---
public_id: valory/ledger:0.19.0
type: connection
config:
  ledger_apis:
    ethereum:
      address: ${SERVICE_HELLO_WORLD_RPC:str:"http://host.docker.internal:8545"}
      chain_id: 31337
      poa_chain: false
      default_gas_price_strategy: eip1559

You should also create a README.md file with the description of your AI agent in plain text or Markdown format:

echo "Your AI agent description." > ./packages/your_name/services/your_ai_agent/README.md

Create an entry for your AI agent in the local registry. Add the corresponding entry to the local registry index file (./packages/packages.json). You must add the entry to the dev section, because it is a component being developed by you. You can use a placeholder for its hash value, as it will be corrected afterwards:
```
{
  "dev": {
    "service/your_name/your_ai_agent/0.1.0": "bafybei0000000000000000000000000000000000000000000000000000",
    /* (1)! */
  },
  "third_party": {
    /* (2)! */
  }
}
```
1. Any other dev entries that you have go here. Entries must be comma-separated (,).
2. Any other third_party entries that you have go here. Entries must be comma-separated (,).
Update the package hashes. The command below will correct any hash mismatch in the service.yaml file, as well as in the local registry index file (./packages/packages.json):
```
autonomy packages lock
```
Perform a sanity-check verification for the AI agent definition. The command below ensures that the AI agent satisfies a number of requirements to be deployed:
```
autonomy analyse service --public-id your_name/your_ai_agent:0.1.0
```
For now, you can ignore any warnings you see. You should see the message:
```
Service is ready to be deployed.
```
Use a local deployment to test the AI agent. This is the recommended approach in order to test your AI agent before you publish it to a remote registry. Follow the instructions in the local deployment guide. Note that this process should be somewhat familiar to you if you have followed the quick start guide.