Atlas FAQs

Question 1

How does Atlas enhance LLM agents?

Accepted Answer

Atlas enhances LLM agents by providing them with highly relevant, structured context via MCP and powerful semantic retrieval using Retrieval-Augmented Generation (RAG). This ensures agents have the precise information needed for complex software tasks.

Question 2

Does Atlas support semantic retrieval and how?

Accepted Answer

Yes, Atlas supports robust semantic retrieval through Retrieval-Augmented Generation (RAG) combined with vectorization technologies like pgvector. This allows for intelligent searching and delivery of relevant information to LLM agents.

Question 3

What is Atlas and its primary purpose?

Accepted Answer

Atlas is a production-grade server that implements the Model Context Protocol (MCP) to provide structured context and semantic retrieval (RAG). Its primary purpose is to enhance LLM agents specifically for software engineering workflows.

Question 4

What are the different context layers Atlas manages?

Accepted Answer

Atlas manages multi-layered context, including Core Context (technology stack, conventions), Workflow Context (feature, bug, or refactor specifics), and Decision Context (approved architectural decisions and ADRs).

Question 5

What is the Model Context Protocol (MCP) that Atlas implements?

Accepted Answer

The Model Context Protocol (MCP) is a standardized protocol that defines how LLM clients request and receive structured context from a server. Atlas implements MCP to provide a consistent and interoperable interface for context management.

Atlas

Atlas

主要功能

使用案例

主要功能

使用案例