Full changelog Object Removal (new) Object Removal allows you to effortlessly remove unwanted objects from your videos: Choose the "Remove" option in the video editing view. Draw a mask around the object you wish to remove (optional) Add a prompt, e.g. “remove the hat”, for
PROMPT ENGINEERING
-
Free ChatGPT Workflow Upgrades: Trigger Words and Memory Prompts
By
–
💡How to upgrade your ChatGPT workflow (for free):
— Futurepedia – Learn to Leverage AI (@futurepedia_io) 3 décembre 2025
❌ Typing a simple question ✅ Using Trigger Words ("Think deeply", "Be thorough")
❌ Scrolling through months of history ✅ Asking Memory Prompts ("What did you tell me about [topic] in our last chat?")
Small tweaks. Massive… pic.twitter.com/TP52VeJ2i5How to upgrade your ChatGPT workflow (for free): Typing a simple question Using Trigger Words ("Think deeply", "Be thorough") Scrolling through months of history Asking Memory Prompts ("What did you tell me about [topic] in our last chat?") Small tweaks. Massive
-
SEO-Optimized Blog Post Generator Prompt
By
–
Steal my Gemini 3.0 prompt to generate SEO-optimized blog posts with meta descriptions, FAQs, and images. —————————–
BLOG POSTS GENERATOR
—————————– Adopt the role of an expert Content Strategist and SEO-Optimized Writer, you're a former -
OpenAI Prepares Memory Search for ChatGPT
By
–
OpenAI might be preparing a Memory Search mode for ChatGPT that will let users pull context directly from it during the conversation. In some form, it is already available as "Browser memory" mode on Atlas, but the label text is different on the web.
-
BrowseSafe Open-Source Tool Hardens Autonomous Agents Against Prompt Injection
By
–
BrowseSafe and BrowseSafe-Bench are fully open-source. Any developer building autonomous agents can immediately harden their systems against prompt injection. Read more:
-
BrowseSafe: Detection Model Against Prompt Injection Attacks
By
–
BrowseSafe is a specialized detection model to defend against evolving prompt injection attacks. It is designed specifically to spot and block malicious instructions hidden in web pages before they can impact AI browser agents.
-
BrowseSafe-Bench: Security Benchmark for AI Browser Agents
By
–
BrowseSafe-Bench is our security benchmark designed to evaluate the robustness of AI browser agents against prompt injection attacks embedded in realistic HTML environments.
-
BrowseSafe: Open-Source Model Prevents Malicious Prompt Injections
By
–
Today we're releasing BrowseSafe and BrowseSafe-Bench: an open-source detection model and benchmark to catch and prevent malicious prompt-injection instructions in real-time.
-
Prompt Injection Attacks: Hidden Malicious Instructions for AI Agents
By
–
Prompt injection involves embedding malicious instructions in text read by AI agents, altering its behavior unnoticed. Attackers hide this in comments, templates, footers, or invisible HTML elements parsed by agents but unseen by users.
-
Extracting Action Items from Meeting Notes
By
–
6/ Meeting notes to action items I paste raw Zoom transcripts into Claude. The prompt: "Extract: (1) all decisions made with owner and deadline, (2) open questions that need answers, (3) action items by person, (4) topics that need follow-up meetings. Format as a structured doc