{"author_url":"https://blog.hatena.ne.jp/m1104m/","provider_name":"Hatena Blog","html":"<iframe src=\"https://hatenablog-parts.com/embed?url=https%3A%2F%2Ftameshitemita.blog%2Fentry%2F2026%2F02%2F26%2F085851\" title=\"OpenClaw\u306e\u30e2\u30c7\u30eb\u3092Codex\u306b\u5909\u66f4\u3059\u308b - \u8a66\u3057\u3066\u307f\u305f\u30d6\u30ed\u30b0\" class=\"embed-card embed-blogcard\" scrolling=\"no\" frameborder=\"0\" style=\"display: block; width: 100%; height: 190px; max-width: 500px; margin: 10px 0px;\"></iframe>","title":"OpenClaw\u306e\u30e2\u30c7\u30eb\u3092Codex\u306b\u5909\u66f4\u3059\u308b","provider_url":"https://hatena.blog","image_url":null,"url":"https://tameshitemita.blog/entry/2026/02/26/085851","categories":["AI","Codex","OpenClaw"],"blog_url":"https://tameshitemita.blog/","height":"190","width":"100%","author_name":"m1104m","description":"\u80cc\u666f \u5148\u65e5gemini API\u306b\u5207\u308a\u66ff\u3048\u305f\u304c\u7279\u306b\u4f55\u3082\u3057\u3066\u3044\u306a\u3044\u304c8\u5186\u307b\u3069\u65e5\u6b21\u3067\u639b\u304b\u3063\u3066\u3044\u305f \u65e2\u5b58\u306e\u30b5\u30d6\u30b9\u30af\u306b\u9069\u7528\u51fa\u6765\u306a\u3044\u304b\u3092\u8003\u3048\u3001ChatGPT\u306eCodex\u3092\u9069\u7528\u3059\u308b\u3053\u3068\u306b\u3057\u305f \u8a66\u3057\u3066\u307f\u305f openclaw onboard --auth-choice openai-codex I understand this is powerful and inherently risky. Continue? yes ChatGPT\u306b\u30ed\u30b0\u30a4\u30f3 .openclaw/openclaw.json \u3092\u7de8\u96c6 \"model\": { \"primary\": \"openai-codex/gpt-5.3-codex\", \"f\u2026","published":"2026-02-26 08:58:51","type":"rich","version":"1.0","blog_title":"\u8a66\u3057\u3066\u307f\u305f\u30d6\u30ed\u30b0"}