How skills and agents work in AWS Quick
Repeatable tasks like status updates, release notes, and ticket creation can be encoded in AWS Quick through agents and skills. This post covers how agents, skills, and your data fit together in AWS Quick, how to build and share skills for your team, and how the official catalog and custom agents help you use them.
AWS Quick is an AI-powered service designed to automate tasks, analyze data, build web applications, and conduct research. It operates in natural language, and utilizes AI agents to process requests against connected data sources and applications.
Agents are a central part of AWS Quick, functioning as configurable AI entities with their own instructions, knowledge sources (spaces), and connectors. These agents ground their responses in data connected to Quick, such as spaces, knowledge bases, datasets, and dashboards. An agent's access is limited to the resources and services the user permits, and is solely determined by the user's permissions.
There are two types of agents in AWS Quick: the default built-in agent, known as Quick, and custom agents, which can be configured by users or team members for specific tasks. Custom agents are set with their own instructions and limited to specific resources, users, and groups within the AWS Quick account.
Data plays a crucial role in AWS Quick. Agents ground their answers in the data connected to the service. A space is a curated collection of documents and Quick resources, which can be shared among teams or groups within the organization. A knowledge base is an organized, indexed collection of documents optimized for retrieval and question answering. It can be synced from various sources like Google Drive, Microsoft OneDrive, and Amazon S3.
Datasets and dashboards, generated by AWS Quick Sight, are available for use in analyses and shared with other Quick users. Connectors link Quick to external applications or native tools, enabling agents to read data and take actions on behalf of users.
Written by urgent.news from AWS Machine Learning's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
This story
This is one outlet's version. Read the fullest account.
