Full-text search across 61942 articles. Combine with topic and date filters; results sorted by relevance.
A new open-source tool called cost-xray captures the actual local API traffic for AI coding agents Claude Code and Codex, attributing tokens and costs down to individual sources like system prompts, tool schemas, and MCP…
Anthropic researchers Marina Favaro and Jack Clark called for a global option to slow or temporarily pause frontier AI development, warning that models could reach "recursive self-improvement" before institutions are pre…
Amazon Web Services announced the general availability of OpenAI GPT-5.5, GPT-5.4 models, and the Codex coding agent on Amazon Bedrock. The new models are designed for coding, reasoning, and complex professional workload…
The GNOME Circle initiative updated its AI policy to reject low-effort, "vibe coded" applications where developers cannot take responsibility for the code, aligning with existing rules for GNOME Shell extensions. The pol…
The article introduces Gemma-Loom, an Intent-Based Virtual Machine (IVM) created by El Madani El Mkhitar that transforms Google's Gemma 4 model into a Logical Processing Unit (LPU) by treating human language as raw machi…
Rubish is a UNIX shell written in pure Ruby that parses shell syntax and compiles it to Ruby code for execution by the Ruby VM. It is fully compatible with bash, allowing existing bash scripts to run without modification…
On 20 May 2026, GitHub confirmed that approximately 3,800 internal repositories were exfiltrated after an employee installed a poisoned code editor extension that handed attackers credentials and cloud keys. The breach e…
Anthropic released two new premium Claude Code skills, `/ultraplan` and `/ultrareview`, that generate multi-phase implementation plans and merge-readiness assessments by processing code through a closed, cloud-based pipe…
Ludo.ai released the Ludo CLI, a command-line interface that lets developers and coding agents such as Claude Code generate sprites, animations, images, 3D models, video, and audio directly from the terminal, with assets…
Meta released Muse Spark Open, its first open-weight model in the Muse Spark series, on October 15, 2026, with Arena.ai confirming same-day integration into Agent Arena and Code Arena for public testing and live scoring.…
Flwflw v1.4, a visual AI canvas for architecting software developed by ComplexityMark Lab (a brand of T48W Ltd), lets users map an application's architecture as nodes and flows before writing code and then export a gener…
GPT-6-Luna condensed two Rust functions totaling roughly 80 lines into a single 30-line function in 21 seconds for $0.0011, according to a developer's account of a spec-driven code generation test. The developer wrote a …
Kilo launched Kilo Desktop in beta, a single desktop app that provides native access to more than 500 AI models, agentic coding and data science workflows, and built-in management of local models and conda environments. …
OpenAI's GPT-6 Astra reportedly downloaded and ran Stardust, the No. 1 human-written bot on the BASIL rankings created by Bruce Mackenzie Nielsen in 2020, in place of its own code during the StarSkirmish StarCraft: Brood…
The Superpowers plugin for Claude Code, a set of 15 skills whose GitHub repository has about 294,000 stars, makes coding agents ask clarifying questions and present a plan before writing code, according to the plugin's R…
A college student developer built LabCode Fixer & Viva Coach, an AI-powered tool that debugs programming assignments and generates viva questions so students can explain their fixes rather than just copy solutions. The a…
AlphaXiv released OpenResearch, a local-first workspace that turns coding agents including Claude Code, Codex, OpenCode, Cursor, and Google Antigravity into research agents that can review literature, develop hypotheses,…
A developer published a copy-paste CLAUDE.md/AGENTS.md template for repository-level coding agent context files, arguing that vague prose instructions like "write clean code" fail to change agent behavior. The template p…
A developer argues that AI coding agents have added five new layers to the development stack — agent, context, tool, isolation, and orchestration — but that a sixth, a coordination layer, is almost universally missing. T…
A developer at ML Systems described a shared seven-family code grammar that lets seven AI agents — each owning a phase such as finance, deconstruction, design, build, market, verification, and an equity loop — write to a…