A developer argues that AI agent skills and plugins are often conflated, but a folder of instructions is not a tool. The developer proposes a clear architecture: skills teach, plugins package, tools a…
A security researcher disclosed a time-of-check-to-time-of-use (TOCTOU) vulnerability in OpenAI's Codex cloud code review that could enable prompt injection attacks. The flaw, which affects the AI-pow…
OpenAI CEO Sam Altman claimed on July 31st that AI cost declines are 20 times faster than Moore's law, after OpenAI announced that its GPT-5.6 Sol model helped cut the cost of serving itself by 20% an…
QM, a multiplayer agent harness for work in Slack and on the web, is designed for startups, allowing each employee to have an isolated workspace while collaborating in channels and projects. The core …
ByteByteGo is hiring a part-time instructor for its live, cohort-based course 'Write Production Grade Code with AI', which teaches software engineers to reliably ship production-grade software using A…
Developer Tanay Kamath has built Nova, an AI-powered personal assistant that offers philosophical life coaching, using LangChain and Mistral AI. The application, hosted on Streamlit, was developed aft…
GoodBarber, an app platform, has deployed a hosted remote MCP server that lets AI agents operate live production apps, including scheduling push notifications and publishing articles. The setup uses O…
DeepSeek released its V4-Flash-0731 API into public beta on July 31, 2026, scoring 82.7 on Terminal Bench 2.1, up from 61.8 in the preview build and surpassing V4-Pro-Preview's 72.1, while maintaining…
Taktile, a software company, has developed an AI coworker named Claudius, powered by Anthropic models, that automates tasks such as investigating tickets, summarizing meetings, and raising pull reques…
A developer used AI coding agents, including Codex powered by GPT-5.4, to rewrite SmallTool, a personal production application maintained for six years, and reported that AI saved significant time and…
DeepSeek has launched the public beta of its V4-Flash API, featuring upgrades for agent tasks, scoring 82.7 on Terminal Bench 2.1 and 54.4 on DeepSWE. The update adds Responses API support and Codex a…
BlackFlare, a new macOS menu bar app, keeps Macs awake during Claude Code and Codex sessions and sends notifications to Mac, Telegram, or Slack when tasks finish or need input. The app, which requires…
Supabase has open-sourced supabase/evals, a benchmark and framework for testing AI agents like Claude Code, Codex, and OpenCode on real Supabase tasks, with results published on supabase.com/evals. In…
DeepSeek released DeepSeek-V4-Flash in public beta on 2026-07-31, showing significantly enhanced agent capabilities with benchmark results far exceeding V4-Pro-Preview, including Terminal Bench 2.1 at…
A developer detailed a fail-closed architecture for autonomous agent networks, introducing a 10-tuple canonical envelope protocol with HMAC-SHA256 witness seals to ensure state consistency across asyn…
A developer ported Cole Medin's agentic coding workflow from Claude Code to OpenAI's Codex, converting Markdown commands into Codex Skills while preserving their semantics as user-initiated workflows.…
OpenAI has cut prices for GPT-5.6 by 20%-80%, with the cost of GPT-5.4-level intelligence dropping 13x in four months due to GPT-5.6's recursive self-optimization. The company published findings showi…
A developer released an AI skill for Codex that filters and reads Hacker Newsletter and other Gmail newsletters, passing raw decoded text MIME parts to the model. The skill, invoked with a date or dat…
OpenAI cut the price of GPT-5.6 Luna by 80% to 20 cents per million input tokens and $1.20 per million output tokens after using GPT-5.6 Soul, running inside its Codex coding agent, to optimize its ow…
A software engineer recounts how mandatory AI training at their company led to the widespread use of large language models to generate business intelligence reports without verification, resulting in …