What does this skill do, and when should you use it?
This skill documents integration testing for Codex's split app-server/exec-server architecture. It covers Linux exec-servers running in Docker and Windows exec-servers running under Wine. Tests currently require an x86_64 Linux host, while macOS developers can use a devbox. The instructions include fixtures for core and app-server tests, configuration-specific skip macros, and runnable Docker and Bazel commands.
It explains how to opt tests into remote execution with TestCodexBuilder::build_with_auto_env() and TestAppServer::new_with_auto_env(); how to start app-server threads with automatic environments; how to select skip_if_target_windows!, skip_if_wine_exec!, skip_if_host_windows!, skip_if_remote!, and skip_if_no_remote_env! based on failure causes; how to source scripts/test-remote-env.sh and clean up Docker environments; how to run codex-core and codex-app-server integration tests; how to run Wine executor tests with Bazel; and how to connect to a macOS devbox over SSH.
- A codex-core maintainer needs to verify integration tests against a Docker remote executor.
- An app-server maintainer needs to test behavior with the app-server and exec-server running separately.
- A cross-platform developer needs to validate a Windows exec-server under Wine.
- A test author needs configuration-specific skips for host, target, Wine, local-only, or remote-only behavior.
- A macOS contributor needs a Linux devbox to run the repository's remote executor tests.
How do you install this skill?
- Remote tests start Docker/Wine executors and may use devboxes over SSH; confirm host permissions, network access, credentials, and test-data boundaries before use.
- Dependency versions, troubleshooting, success criteria, and behavior outside the stated Linux/x86_64 target are not fully documented.
- Shell / CLI
- Network access
- Local filesystem
x86_64 Linux hostDockerWineBazelBashSSH for macOS devboxes
The README documents installation of the overall Codex CLI collection, including curl -fsSL https://chatgpt.com/codex/install.sh | sh on macOS or Linux, npm install -g @openai/codex, and brew install --cask codex. It does not document a separate installation or registration procedure for .codex/skills/remote-tests/SKILL.md.
tmp="$(mktemp -d)"
git clone --depth 1 https://github.com/openai/codex.git "$tmp"
mkdir -p ~/.claude/skills
cp -R "$tmp/.codex/skills/remote-tests" ~/.claude/skills/
rm -rf "$tmp"Generated from the source repository and skill path; it copies only this skill's folder. If the author's install steps above differ, follow those first. To scope it to one project, replace ~/.claude/skills with that project's .claude/skills.
How do you use this skill?
Once installed, send your agent any of these to trigger it:
- set -euo pipefail; unset CODEX_TEST_REMOTE_EXEC_SERVER_URL; source scripts/test-remote-env.sh; trap codex_remote_env_cleanup EXIT; cd codex-rs; just test -p codex-core --test all
In a Codex checkout, opt relevant tests into automatic remote environments. For Docker core tests, run: bash -c 'set -euo pipefail; unset CODEX_TEST_REMOTE_EXEC_SERVER_URL; source scripts/test-remote-env.sh; trap codex_remote_env_cleanup EXIT; cd codex-rs; just test -p codex-core --test all'. Replace the package with codex-app-server for app-server tests. For Wine, run bazel test //codex-rs/core:core-all-wine-exec-test or bazel test //codex-rs/app-server:app-server-all-wine-exec-test.
What are this skill's strengths and limitations?
- Covers both Docker Linux exec-server and Wine Windows exec-server configurations.
- Provides concrete fixture patterns and configuration-specific skip macros.
- Includes copyable Docker, Bazel, and devbox workflows.
- Remote executor tests currently require an x86_64 Linux host machine.
- Wine tests run only in Bazel because of the cross-platform build dependency.
- The source does not specify a standalone skill installation process or complete environment version matrix.
How does this skill compare with similar options?
Side by side with related skills; every score comes from the same FSRS standard.
| Skill | FS score | Stars | Last updated | License |
|---|---|---|---|---|
| Remote Executor Integration Tests this page ✓ OpenAI · Official | 50 · Use with care | ★ 129k | 11d ago | Apache-2.0 |
| Marmite Development Guide | 58 · Recommended | ★ 886 | 26d ago | AGPL-3.0 |
| Code Review and Testing Guidance ✓ OpenAI · Official | 43 · Not recommended | ★ 129k | 11d ago | Apache-2.0 |
| Optional-Dependency Smoke Tests (agent-service-toolkit) | 64 · Recommended | ★ 4.5k | 21d ago | MIT |
| NeMo Relay Installation Guide ✓ NVIDIA · Official | 62 · Recommended | ★ 3.6k | 11d ago | Apache-2.0 |
How did FollowSkills review this skill?
The skill scopes itself to remote-executor testing and discloses Docker, Wine, environment-variable, and devbox flows; however, it does not explain sensitive-data handling, privilege boundaries, user confirmation, rollback, or the trust model for remote hosts, so points are deducted.
It gives concrete commands and skip macros for codex_core, app-server, Docker, and Wine, making the main paths understandable; dependency requirements, version compatibility, failure diagnosis, and abnormal-input handling are thin, and static review cannot establish executability, so points are deducted.
The audience and scenarios are clear, covering Linux, Wine, and macOS devboxes; however, the x86_64 Linux requirement, non-fit boundaries, trigger conditions, and Chinese-environment support are under-specified, and remote/devbox access may be affected by mainland-China network reachability, so points are deducted.
The document is concise and organized around fixtures, skip rules, Docker, Wine, and devbox examples, with Apache-2.0 licensing and OpenAI provenance available in repository context; installation dependencies, versioning, changelog, maintenance ownership, and troubleshooting are missing, so points are deducted.
It provides directly usable commands for core and app-server remote tests and addresses the primary task; expected outputs, success criteria, alternatives, and actionable failure results are insufficiently documented, and static evidence cannot confirm actual effectiveness, so points are deducted.
Specific commands, macro names, and script paths provide limited auditability, and repository workflow context mentions Linux remote-environment tests; there is no skill-specific reproducible execution evidence or independent corroboration, and nothing was executed in this review, so points are deducted.
Open a dimension to read why it scored that way
Evidence confidence：Low — Mostly static review, author material or a limited demo; useful for discovery, not high-risk decisions.
See the full review method →