Search for a command to run...
On 12.7% of the AI agent skills that two auditors both rated, one says HIGH or CRITICAL while the other says LOW or SAFE about the exact same code. That number comes from the Agent Skill Security Cens
Disclosure: I run Skillselion, a live catalog of agent skills refreshed daily from skills.sh, GitHub and MCP registries, ranked by real installs. Every number below was pulled from it on August 12, 20
On July 17, Cursor added plan-first Slack replies, named multi-repository environments, and the ability for its agent to read and post across channels. Once a coding agent can start from Slack, coordi
Open the Claude Code changelog from this week and read it the way you would read a CVE feed. Version 2.1.212 fixed plan mode auto-running file-modifying Bash commands like rm without a permission prom
Pull the Google Trends breakout list for Claude Code skills this week and the demand does not match the discourse. The loud posts are still about the 132-agent mega-stack: dozens of subagents, a wall
mattpocock/skills is the most-installed personal skills repo in the Claude Code ecosystem. By Matt Pocock's own count: "160K stars, 7.5m downloads... and no tutorial" (Jul 9, 2026). It is also unusual
Yesterday Anthropic shipped a sandboxed browser inside the Claude Code desktop app. Cmd+Shift+B, and your agent can read docs, click through a signup flow, and poke at your local dev server without ev
OpenAI shipped the GPT-5.6 family (Sol, Terra, Luna) yesterday and folded Codex into a unified ChatGPT desktop app. The loudest reaction on X was not about benchmarks. It was a developer on the $200/m
It is 1 AM and you need a Postgres MCP server. Not a list of Postgres MCP servers: the one you should actually run. You check five directories and get five different answers. One sorts alphabetically,
