Full-text search across 65714 articles. Combine with topic and date filters; results sorted by relevance.
Perplexity has launched SPACE (Sandboxed Platform for Agentic Code Execution), a runtime for long-running AI agents that execute code and complete multi-step tasks over extended periods. The system, which powers 100% of …
Beijing-based startup Moonshot AI released Kimi K3, a 2.8-trillion-parameter open-weight model with a 1-million-token context window, claiming it matches or surpasses top-tier models from OpenAI and Anthropic on coding a…
A cybersecurity lecturer, Katie Paxton-Fear, demonstrated that for less than $100 and an hour of work, she could sneak a backdoor into an open-weight AI model to enable remote code execution, highlighting vulnerabilities…
Microsoft has open-sourced Comic Chat, a 1996 IRC chat client with comic-style avatars, after 30 years. The company released the source code along with AI-powered modernization attempts to make the 1990s-era C++ code bui…
A developer demonstrated how to separate only CUDA kernels into .cu files while keeping the rest of the code in .cpp files, using a minimal noopKernel example. The approach confines CUDA-specific code to a limited set of…
A developer building a Chrome extension that detects HTML code blocks inside ChatGPT, Claude, and Gemini and deploys them to live URLs found that reliably watching the constantly-changing chat UIs was the hardest part. K…
A developer built a fully automated coding loop in OpenCode, achieving surprisingly good results with the concept of Loop Engineering. The project demonstrates an automated workflow for coding tasks using AI.
Quantum Circuit Vision, a new benchmark of 132 quantum circuits across 13 categories, evaluates multimodal AI agents' ability to understand circuit diagrams and generate executable code for Amazon Braket. Anthropic's Cla…
OpenAI on Wednesday released the Codex Micro, a $230 miniature keyboard for monitoring and controlling Codex AI coding agents, marking the company's first consumer hardware device. Developed with peripheral maker Work Lo…
Zapier launched the Zapier Marketplace, a plugin system that lets AI coding agents like Claude Code, GitHub Copilot CLI, and OpenAI Codex connect to thousands of apps including Notion, Google Calendar, and Slack directly…
A new study from arXiv:2607.12161v1 analyzing 2,848 Claude Code runs across 103 tasks finds that reducing retrieved context or tool output does not reliably lower billed costs for API-based coding agents, and can harm ta…
A security researcher published a wire-level analysis on July 12 proving that xAI's Grok Build coding CLI was packaging developers' entire tracked repositories, including full Git history, committed secrets, and API keys…
Amazon SageMaker AI launched a UI for generative AI inference recommendations in SageMaker AI Studio, providing a low-code no-code experience that guides users through preset use-case profiles, visual comparisons, and on…
Large language models have advanced to the point where they can solve open mathematical problems, generate accurate QR codes, and handle real-time customer service calls, according to a compilation of examples from resea…
A developer shared a starter configuration for Claude Code, a tool that uses AI to assist with coding tasks. The configuration includes a curated allowlist of safe commands, integration with the open-source RTK tool to r…
Rapid7's Metasploit Framework has added a new exploit for FlowiseAI Flowise (CVE-2026-41264), an unauthenticated remote code execution vulnerability in the CSV Agent feature that allows attackers to upload a .csv file co…
Researchers introduced ANYPoC, a multi-agent framework that automates bug validation by generating executable proofs of concept. The system outperformed state-of-the-art coding agents like Claude Code and Codex, producin…
Researchers present a harness-engineering approach for building auditable enterprise LLM agents, moving deterministic behavior into code and validation artifacts. Tested on Korean corporate data, the harness preserved so…
As of Claude Code 2.1.201 in July 2026, Anthropic's tool still does not natively read AGENTS.md, forcing developers to use a symbolic link workaround. The distinction between CLAUDE.md, AGENTS.md, and SKILL.md lies in th…
Telnyx published a Python example that builds a CDR usage analytics dashboard using Flask and Telnyx AI Inference. The dashboard provides routes for metrics like summary, peak hours, and top routes, and uses AI to genera…