---
variant: detail
kind: decisions
slug: use-codex-and-claude-apps-as-primary-engineering-environment
url: /decisions/use-codex-and-claude-apps-as-primary-engineering-environment
title: Use Codex and Claude macOS apps as my primary engineering environment
source_path: >-
  content/decisions/use-codex-and-claude-apps-as-primary-engineering-environment.md
frontmatter:
  decision: Use Codex and Claude macOS apps as my primary engineering environment
  date: '2026-05-14T00:00:00.000Z'
  status: standing
  context: >-
    My day-to-day software work has moved from an editor-centered workflow into
    agent desktop apps that hold the repo, shell, browser, instructions, and
    review loop.
  options_considered:
    - >-
      Keep VS Code as the primary engineering environment and use agent apps as
      assistants.
    - >-
      Use Codex and Claude as the primary engineering environment, with VS Code
      reserved for direct file-oriented tasks.
    - 'Split primary work evenly between VS Code, Codex, and Claude.'
  chosen: >-
    Default to agent workbenches for engineering work, with Codex first when
    browser-backed inspection matters and VS Code reserved for precise file
    edits.
  why: >-
    Most engineering work now benefits more from an agentic workbench that can
    coordinate repository context, terminal commands, browser inspection, edits,
    and validation than from direct manipulation inside a traditional editor.
  what_would_change_my_mind: >-
    I would reconsider if agent desktop apps became less reliable for repository
    work, if their browser and shell loops stopped feeling integrated, or if VS
    Code regained a clear advantage for the majority of engineering tasks.
  reverses: []
  related_claims:
    - claims/agent-harness-absorbs-complexity
  related_projects:
    - projects/agent-harness
  tags:
    - agents
    - tools
    - development
links:
  related_claims:
    - status: resolved
      ref: claims/agent-harness-absorbs-complexity
      kind: claims
      slug: agent-harness-absorbs-complexity
      title: >-
        Agent harnesses absorb complexity that would otherwise require separate
        infrastructure scaffolding such as service meshes or orchestration
        engines.
      url: /claims/agent-harness-absorbs-complexity
  related_projects:
    - status: resolved
      ref: projects/agent-harness
      kind: projects
      slug: agent-harness
      title: Agent Harness
      url: /projects/agent-harness
agent_metadata:
  source_path: >-
    content/decisions/use-codex-and-claude-apps-as-primary-engineering-environment.md
  html_url: /decisions/use-codex-and-claude-apps-as-primary-engineering-environment
  markdown_url: /decisions/use-codex-and-claude-apps-as-primary-engineering-environment.md
  source_url: >-
    https://github.com/flaming-codes/thinkinglabs/blob/main/content/decisions/use-codex-and-claude-apps-as-primary-engineering-environment.md
  summary: Codex and Claude's macOS apps are now my primary engineering environment.
  word_count: 471
  approx_token_count: 875
  token_estimate: chars/4
---
Codex and Claude's macOS apps are now my primary engineering environment.

VS Code is still part of the loop, but mostly for direct file-related tasks: quick manual inspection, targeted edits, or work where I specifically want an editor surface. The center of gravity has shifted elsewhere.

For most engineering work, I now prefer to start in Codex or Claude. Codex is the default when the task benefits from the immersive browser view, because it makes implementation, local verification, and UI inspection feel like one continuous workspace rather than separate tools I have to manually coordinate. Claude remains useful as another high-quality agentic engineering surface, especially when I want a different model's judgment or interaction style.

The decision is less about brand loyalty than workflow shape. The thing I want to inhabit is no longer just a file tree plus terminal. It is the loop that moves from intent to context gathering, patch, test, browser inspection, correction, and handoff.