Skip to content

Error Codes and Exit Codes

Atlas uses structured error reporting to keep failures classifiable and automatable.

Error Reporting Model

flowchart LR
    Failure[Failure] --> ApiError[API error code]
    Failure --> ExitCode[CLI exit behavior]
    ApiError --> Client[Client handling]
    ExitCode --> Automation[Automation handling]

This error-reporting model separates two important consumers: HTTP clients and CLI automation. Both need structured signals, but they consume them through different surfaces.

Main API Error Categories

flowchart TD
    Errors[API error categories] --> Query[Query and validation]
    Errors --> Auth[Authentication and access]
    Errors --> Data[Dataset and artifact]
    Errors --> Runtime[Runtime and readiness]

This category map helps readers interpret error codes by failure area instead of memorizing each one in isolation.

Common API Error Classes

  • authentication and access errors
  • invalid query or missing dataset dimension errors
  • dataset or gene not found errors
  • policy rejection and cost rejection errors
  • readiness, timeout, and upstream availability errors
  • artifact corruption or quarantine errors

Important Point

Use the structured error code, not only the human message text, when building client or automation behavior.

Purpose

This page is the lookup reference for error codes and exit codes. Use it when you need the current checked-in surface quickly and without extra narrative.

Stability

This page is a checked-in reference surface. Keep it synchronized with the repository state and generated evidence it summarizes.