About
Atlas is built for engineers who are tired of re-explaining the repo.
The product goal is narrow: give AI coding agents a persistent, cited, local understanding of the codebase without uploading repository contents during indexing.
Principles
Small surface, serious claims.
Atlas is not trying to become the coding agent. It is the memory layer those agents can query when they need grounded context.
- 01
Local first
Repository scanning and indexing happen on your machine.
- 02
Cited by default
Answers point back to files and evidence, so engineers can verify before acting.
- 03
Agent neutral
The same local index can serve Claude Code, Cursor and Codex over MCP.
Limits
The constraints are public too.
Numbers are self-measured on one Windows development machine; your hardware will differ.
The 50-scenario suite runs against a reference repository we built, not a random sample of OSS projects.
Large monorepo indexing is slower and is an active work item.
Retrieval quality is not the same as end-to-end task success with a coding agent.