Google DeepMind expands AI safety partnership with Singapore Anthropic finds over 10,000 vulnerabilities with Project Glasswing SynthID expands to Google Search and Chrome Anthropic updates vuln disclosure dashboard Goal mode now available across all Codex platforms Codex Thursday adds remote Mac control Anthropic shares early Glasswing results Anthropic publishes early Project Glasswing results Releases new science-focused AI skills tool Gemini 3.5 Flash released with enhanced research tools Google ships ADK for Android/Kotlin v0.1.0 Google launches ADK for Kotlin and Android 0.1.0 Google expands Gemini for Home for developers Gemini 3.5 Flash officially launched AI solves long-standing open math problem for first time Google announces Gemini Omni for video creation Use multiple agents with Gemini Omni OpenAI Introduces Guaranteed Capacity for Long-Term Compute Gemini for Science assists with research tasks SynthID watermark and verification tool added to AI images Google DeepMind expands AI safety partnership with Singapore Anthropic finds over 10,000 vulnerabilities with Project Glasswing SynthID expands to Google Search and Chrome Anthropic updates vuln disclosure dashboard Goal mode now available across all Codex platforms Codex Thursday adds remote Mac control Anthropic shares early Glasswing results Anthropic publishes early Project Glasswing results Releases new science-focused AI skills tool Gemini 3.5 Flash released with enhanced research tools Google ships ADK for Android/Kotlin v0.1.0 Google launches ADK for Kotlin and Android 0.1.0 Google expands Gemini for Home for developers Gemini 3.5 Flash officially launched AI solves long-standing open math problem for first time Google announces Gemini Omni for video creation Use multiple agents with Gemini Omni OpenAI Introduces Guaranteed Capacity for Long-Term Compute Gemini for Science assists with research tasks SynthID watermark and verification tool added to AI images

🔒 公式発表のみ掲載。噂・リーク・情報商材は載せません。

← Back to top

OpenAIGuides & TipsOfficial Docs

OpenAI shares operational tips in Codex Prompting Guide

Reduces agent failures while cutting wasted reads and tokens.

Key Points

1Templates for parallel tool-calling
2Recommends ~10k-token tool output truncation
3Explains phase handling for long tasks
4Batch file reads instead of one-by-one

OpenAI’s Codex Prompting Guide explains how to write instructions that work reliably for coding agents. It details parallel tool-calling workflows and how to truncate tool outputs to save tokens. It also notes implementation cautions like the phase parameter for longer tasks. Existing Codex integrations may benefit from revisiting state and parallelism assumptions.

📎 Source: Official Docs