Codex Computer Use

Optional cross-vendor executor. Requires the OpenAI codex CLI. If command -v codex fails, this skill isn't available on this machine — say so and verify by another means. The model-orchestration doctrine (references/model-orchestration.md) routes here only when the CLI is present.

Use Codex as a separate local verification agent when the task needs real UI interaction, screenshots, simulator/browser/device state, or an independent runtime check outside Claude's current context.

Do not use this for ordinary code reading, typechecking, linting, or tests Claude can run directly. Launching apps, simulators, or browsers to verify the requested work is fine without asking; ask first only if the run could disrupt the user's environment beyond that (closing their apps, changing system settings, acting on real accounts or data).

Workflow

Define exactly what behavior to verify and on what platform.
Create a temporary artifact directory for screenshots, logs, and the report.
Write a self-contained prompt and run codex exec with the sandbox level the task needs.
Read the report and screenshots, and verify key claims before presenting them.

ARTIFACT_DIR="$(mktemp -d "${TMPDIR:-/tmp}/codex-computer-use.XXXXXX")"
REPORT="$ARTIFACT_DIR/report.md"
PROMPT="$ARTIFACT_DIR/prompt.md"