What does this skill do, and when should you use it?
This skill guides interactive testing of the Codex TUI. It requires starting the process interactively with trace-level logging enabled. Test runs should write logs to a specified temporary directory, and programmatic input must send the message text and Enter as separate writes. It is best suited to developers working in the Codex repository who need to verify terminal-interface behavior or investigate failures.
It instructs the user to launch Codex TUI with the repository's just codex target, set RUST_LOG="trace", pass -c log_dir=<some_temp_dir> for log output, and send programmatic test input as two operations: the text first, followed by a separate Enter write.
- A Codex CLI developer has changed the TUI and needs to verify its interactive behavior.
- A tester needs to reproduce a terminal interaction issue while preserving trace-level logs.
- A developer is sending test messages programmatically and must avoid combining the text and Enter keystroke.
- A Codex repository maintainer needs the documented `just codex` entry point for running the TUI.
How do you install this skill?
- Before use, confirm TUI permissions, log-directory scope, and whether logs may contain prompts, source code, or credentials.
- The skill defines no failure criteria, expected outputs, or cleanup and rollback procedure; successful startup alone does not establish that a change was correctly verified.
- There is no evidence of Chinese-language support or mainland-China network reachability; availability may be constrained if login or service access depends on overseas connectivity.
- Shell / CLI
- Local filesystem
justCodex CLI
Place the skill directory where the Agent Skills client discovers skills, with the file at .codex/skills/test-tui/SKILL.md. The source does not document a universal installation path or a standalone installation command. The test environment also needs the Codex CLI and the repository's just codex target.
tmp="$(mktemp -d)"
git clone --depth 1 https://github.com/openai/codex.git "$tmp"
mkdir -p ~/.claude/skills
cp -R "$tmp/.codex/skills/test-tui" ~/.claude/skills/
rm -rf "$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?
From a Codex repository development environment, run RUST_LOG="trace" just codex -c log_dir=<some_temp_dir>. Start the test interactively. For programmatic input, write the test message first and then issue a separate write containing Enter. The source does not prescribe a fixed test prompt.
What are this skill's strengths and limitations?
- Defines a precise launch command pattern and logging configuration.
- Clarifies the interactive startup requirement and separate Enter write.
- Uses a dedicated log directory to support debugging.
- Covers Codex TUI testing guidance rather than a general-purpose TUI test framework.
- Depends on the `just codex` target and a working Codex CLI environment.
- Provides no test suite, assertion strategy, or failure-triage procedure.
- Does not document testing steps for Windows or other platforms.
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 |
|---|---|---|---|---|
| Codex TUI Interactive Testing Guide this page ✓ OpenAI · Official | 37 · Not recommended | ★ 129k | 11d ago | Apache-2.0 |
| debug-skill: Interactive Debugger for Coding Agents | 56 · Use with care | ★ 331 | 6mo ago | MIT |
| BlockWatch | 67 · Recommended | ★ 29 | 11d ago | MIT |
| NeMo Relay Installation Guide ✓ NVIDIA · Official | 62 · Recommended | ★ 3.6k | 11d ago | Apache-2.0 |
| NeMo Relay Quick Start ✓ NVIDIA · Official | 61 · Recommended | ★ 3.6k | 11d ago | Apache-2.0 |
How did FollowSkills review this skill?
The skill requires interactive startup, trace logging, and a specified temporary log directory, which improves debugging visibility. It does not define permissions, sensitive-data or log handling, network and external effects, user confirmation, isolation boundaries, or rollback, so points are deducted.
The short procedure is internally consistent and specifies the just codex target, logging argument, and separate text and Enter writes. However, it assumes just, Codex TUI, and a suitable terminal environment, with no diagnosis or recovery guidance for startup failures, invalid parameters, missing tools, or abnormal input; the static cap therefore applies and points are deducted.
The intended scenario is interactive verification of Codex TUI changes, and the basic audience and trigger are clear. Non-fit boundaries, input/output boundaries, platform requirements, Chinese-language support, and mainland-China network fit are not stated or evidenced, so points are deducted.
The file has basic name, description, and operational notes, while repository context supplies Apache-2.0 licensing and OpenAI attribution. It lacks installation prerequisites, parameter detail, examples, FAQs, known limitations, versioning, changelog, maintenance ownership, and update path, so points are deducted.
Following the instructions can support starting the TUI, retaining trace logs, and performing programmatic input, covering the core verification action. Expected results, a validation checklist, output format, and a complete failure workflow are unspecified, leaving substantial manual interpretation; only limited points are awarded.
The skill text is auditable, and the repository context includes related tests and CI material. Those files do not demonstrate execution of this skill or coverage of its key paths, and no third-party execution evidence is provided, so static verifiability is limited and 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 →