Lelu: The New Tool to Catch AI Agents When They're Manipulated
Lelu is an open-source tool designed to detect and prevent manipulation of AI agents in real-time. It's a significant step forward in ensuring the security and reliability of AI systems.
#tools
Tools
262 stories tagged Tools · page 6 of 11
AuraText Brings AI Suggestions to Any Text Field on Windows
AuraText is a new Windows overlay that adds AI-powered text suggestions to any text field. It works with all applications, making it easier to write and edit text on your computer. This tool could be a game-changer for anyone who types a lot, from students to professionals.
Hugging Face Releases Tool to Benchmark AI Agent Performance
Hugging Face introduced a new tool to test how well open-source AI models perform as agents. This helps developers compare models without needing complex setups.
OSymandias: Open-source runtime for multi-agent AI systems
OSymandias is a new open-source runtime that lets developers build and run AI systems with multiple agents working together. It simplifies creating complex AI workflows that previously required expensive tools.
Khwand: Self-Healing CI/CD for AI Agents Launches on GitHub
Khwand is a new GitHub app that automates testing and fixing AI agents. It promises to make developing AI-powered tools more reliable for developers.
Retool Launches React AI App Builder
Retool has introduced an AI tool that lets anyone build React apps without coding. This could make custom business tools much more accessible to small teams and startups.
GLM-5.2: Open-Source AI Model Excels at Complex, Multi-Step Tasks
GLM-5.2 is a new open-source AI model designed to handle long-horizon tasks better than most. It's now available for anyone to use and build upon.
Developers Share Favorite AI Models for Everyday Features
A Hacker News thread reveals how developers choose AI models for daily features like content generation and recommendations. Many focus on cost, speed, and ease of use over cutting-edge performance.
File Systems Are the New Primitive for AI Agents
A new wave of AI agents is emerging that uses file systems as their core primitive for managing tasks and data. This architectural shift could make agents more capable, transparent, and practical for everyday users.
Why Hollywood Isn't Relying on Generic AI for Film Making
Generic AI models aren't cutting it for Hollywood. The industry needs specialized tools to create engaging content. This shift could change how movies and shows are made in the future.
Allen Institute Releases Olmo-Eval: Open-Source Tool for AI Model Testing
Olmo-Eval is a new open-source platform designed to help developers test and improve AI models. It streamlines the evaluation process, making it easier for anyone to assess AI performance.
New Diagnostic Framework ToolSense Helps LLMs Find the Right Tools Faster
Researchers introduced ToolSense, a diagnostic framework for auditing parametric tool knowledge in LLMs. It helps AI agents select tools more accurately by encoding each tool as a virtual token and fine-tuning the model in two stages, outperforming traditional embedding-based retrieval on standard benchmarks.
AfriSUD: A New Collection of Syntactic Treebanks to Help AI Understand African Languages
Researchers created AfriSUD, the first large-scale collection of syntactically annotated treebanks for nine diverse African languages, verified by native speakers. This is a major step toward making NLP more inclusive for millions of speakers across Sub-Saharan Africa.
AI Mediator Preps Negotiations Before You Even Talk
Researchers developed an AI system that helps people prepare for negotiations by analyzing their goals and strategies. This could make negotiations faster, fairer, and more effective for everyone involved.
Malware Developers Exploit AI Safety Measures to Bypass Detection
Malware developers are adding text about nuclear and bioweapons to their code to trigger AI safety refusals. This tactic allows them to bypass detection mechanisms in AI systems designed to prevent harmful content.
NotebookLM gets Gemini 3.5 upgrade with cloud computing and source-finding tools
Google's NotebookLM app now uses the more powerful Gemini 3.5 model, offering better accuracy and a new cloud computer feature. The upgrade also helps find sources, but the cloud computer is a separate, related tool for computation.
KiloCode's AI Toolkit Reaches 100,000 Users in First Month
KiloCode, a startup offering AI tools for creators, hit 100,000 users in just one month. The platform provides easy-to-use AI tools for content creation, design, and automation, making advanced AI accessible to everyone.
AI-Powered Tool Lets Anyone Create Custom Chatbots in Minutes
A new no-code platform called Botify.ai lets users build their own AI chatbots without coding. This could make AI tools more accessible to small businesses and hobbyists.
![ReARM: Governing AI Coding Agents Demo [video]](/images/articles/thumbs/2026-06-08-new-ai-system-rearm-demonstrates-how-to-govern-ai-coding-agents.webp?dpl=dpl_FG4W6Fc9AeFEFH6KxnhoUrCjPXPz)
ReARM: Governing AI Coding Agents Demo [video]
A new system called ReARM demonstrates how to govern AI coding agents, ensuring they follow rules and perform tasks safely. Watch the demo video to see it in action.
Codex is becoming a productivity tool for everyone
OpenAI's Codex is evolving beyond coding to help with research, data analysis, and content creation. It's making complex tasks easier for everyday users, not just programmers.
EVA-Bench Data 2.0: The New Benchmark for AI Tool Use
EVA-Bench Data 2.0 is a comprehensive dataset designed to test AI models' ability to use tools effectively. It includes 213 scenarios across 3 domains and 121 tools, making it a valuable resource for developers and researchers.
Microsoft's New Tool Lets Developers Test AI Behavior with Text Descriptions
Microsoft released an open-source tool that lets developers create AI behavior tests using simple text descriptions. This could make it easier to ensure AI systems behave as intended.
OpenAI's new Codex tools bring AI to white-collar jobs
OpenAI released six specialized plug-ins for its Codex app, each targeting a specific white‑collar role: data analytics, creative production, sales, product design, equity investing, and investment banking. The tools bundle integrations, instructions, and context so Codex can approximate human‑level job performance.
When Helping Hurts: How Multi-Agent AI Debate Can Improve—or Degrade—Data Cleaning
New research from MIT and Stanford shows that multi-agent AI debate can significantly boost data-cleaning accuracy in some contexts while introducing harmful errors in others. The key insight: debate improves error detection but can degrade data generation. Here's what to watch for.