The New ChatGPT Playbook & Prompting Guide

The New ChatGPT Playbook & Prompting Guide

by Matt Wolfe

Applied Sciences & Engineering AI — Models & Research UYQArtificial intelligence (AI)UYQFGenerative artificial intelligence
🎙 Matt Wolfe 👥 1.0M 📅 July 15, 2026 ⏱ 38 min 👁 67K 📄 tutorial 🧭 2026-08-28
Available in: English (current) Français

Keywords

ChatGPT 5.6AI assistantprompt engineeringCodexautomation

Summary

Matt Wolfe presents a comprehensive guide to the new ChatGPT app, which integrates ChatGPT, Codex, a browser, and ChatGPT Work into a single platform. He begins with a tour of the app’s features, including the new Sites feature for easy web publishing, a built-in browser with annotation capabilities, and scheduled tasks for automation. The core of the video is a series of prompt demonstrations, starting with setting up a personal assistant that can access all connected apps and data. He shows how to create a book one-pager, generate slide decks, research using a logged-in browser, and prepare for meetings or panels. He also demonstrates using ChatGPT Work to clean up folders and rename files. The video then shifts to Codex, where he showcases the new 5.6 models’ ability to build websites, games, and even control Blender. He highlights community creations like AI-generated videos and 3D city builds. Finally, he shares his own ‘Wolf Control Tower’ project and one-prompt mini apps, concluding with final thoughts on the platform’s capabilities.

168 words

Critical Evaluation

Value of the Information & Strength of the Argument

The video provides high practical value, offering concrete, actionable prompts and workflows that viewers can adapt. The demonstrations are clear and effective, showing real-world applications of the new features. The argumentation is based on the creator’s extensive personal experience, which lends credibility. However, the sponsorship by OpenAI is a potential conflict of interest, though the creator explicitly states his commitment to honesty. The video is more of a showcase than a critical evaluation, but it does not shy away from mentioning limitations, such as the need to review AI-generated content.

Scientific Rigor, Source Quality, Title Accuracy

The video is a tutorial based on the creator’s direct experience, not a scientific review. It does not cite external sources, but the description provides links to the ChatGPT download, Future Tools, and social media. The title accurately reflects the content. The sponsorship is clearly disclosed, which is a positive sign for transparency. The video’s rigor is limited by its promotional nature, but the demonstrations appear genuine and the creator’s expertise is evident.

178 words

Title / Content Match

The title accurately reflects the content: a comprehensive guide to the new ChatGPT app and prompting techniques.

Quality & Reliability

7/10

The video is a practical tutorial based on the creator's extensive hands-on experience (6.8 billion tokens used). It clearly discloses sponsorship by OpenAI, which introduces a potential bias, but the creator emphasizes honesty and provides concrete demonstrations. The information is current and detailed, though not peer-reviewed.

Chapters

Cited Sources

  • ChatGPT Download — Official download link for the new ChatGPT app, mentioned in the video description.
  • Future Tools — Matt Wolfe's website for AI tools and news, referenced in the video.
  • Future Tools Newsletter — Weekly newsletter from Future Tools, mentioned in the video description.
  • Matt Wolfe on LinkedIn — Matt Wolfe's LinkedIn profile, linked in the video description.
  • Matt Wolfe on Threads — Matt Wolfe's Threads profile, linked in the video description.

Concurring Sources

  • OpenAI Codex — Official page for OpenAI's coding agent, which is a key component of the new ChatGPT app.

Contribution & Novelties

The video offers a timely, hands-on overview of the newly integrated ChatGPT app, showcasing features like the Sites hosting, browser annotation, and the Work mode’s personal assistant capabilities. It provides a collection of practical prompts that go beyond basic usage, demonstrating how to leverage the model’s access to personal data and connected apps for real productivity gains. The ‘Wolf Control Tower’ example illustrates a sophisticated, personalized automation setup.

Pour aller plus loin :

  • OpenAI Codex — Official page for OpenAI’s coding agent, relevant to the Codex section.
  • Prompt engineering — Wikipedia article on the practice of designing prompts for LLMs, directly relevant to the video’s core topic.
  • AI agent — Wikipedia article on intelligent agents, which underpins the ‘Work’ mode’s autonomous task execution.

123 words

Radar Profile

The radar profile shows high scores in information quantity and quality, reflecting the video's comprehensive and practical content. The technical level is moderate, making it accessible to a broad audience. The global reliability is slightly lower due to the sponsored nature and lack of external citations.

Reliability 7/10

💬 Très positif. Sur les 30 commentaires analysés, la grande majorité exprime une forte appréciation, saluant la qualité du guide, son utilité pratique et la transparence de Matt Wolfe concernant le sponsoring.