Smart Glasses Skills: teaching coding agents to build for every pair of glasses
Agent skills for Even Realities G2, Meta Ray-Ban Display, Rokid, Vuzix and more, with a source and a date for every fact.
Security advisories, builder stories, and engineering notes on AI agents, MCP, and the open standards behind them.
245 articlesRSS
Agent skills for Even Realities G2, Meta Ray-Ban Display, Rokid, Vuzix and more, with a source and a date for every fact.
The best ChatGPT plugins in 2026: 10 picks that run on a hosted server, so they work on chatgpt.com, with HOL scan results, how to add each one, example prompts, and pros and cons.
Claude Code ran supabase db reset to fix a failed migration. HOL Guard paused it before your local data was wiped: Allow just this once or Keep blocked.
Handoff uses a Markdown file to carry decisions, constraints, and next steps between AI coding sessions. Its creator explains the workflows and how to get started.
I built humanizer-lint to flag stock phrases and repetitive rhythm in drafts without rewriting the text.
A Claude Code and Codex plugin that checks an agent's "done" against a fresh verification record before the turn ends, built from a public ledger of real defects.
Metamynd authorization for NVIDIA OpenShell, shown as a proof of concept on one host: a purchase from a sandboxed agent is checked against its live mandate, OpenShell swaps in the real credential, and the gateway settles it.
Research Agent Skills helps AI agents support academic research with 280+ skills for literature reviews, experiments, data analysis, writing, citations, and peer review.
Mnemosyne OS is a desktop control room for your AI agents: status cards, reviews from your phone, and a memory that stays on your disk. MIT MCP server.
An MCP server between your AI app and its tools: every call checked before it runs, and the API keys kept in the gateway so the AI can't go around it.
Env-guard checks Claude Code's Read, Bash and Grep requests for .env paths and mobile signing keys. The maintainer explains setup, tracked-file checks and the guard's limits.
The best Antigravity plugins in 2026: 10 picks for Google's Antigravity CLI, sorted by what still works after agy plugin install, with HOL scan results, install commands, example prompts, and pros and cons.
Wrap an AI agent's tools in three lines: every call is decided outside the agent against a mandate it can't edit, and a refused call never runs.
Threat guides and evidence dossiers on prompt injection, MCP tool poisoning, slopsquatting, and the attacks shaping how teams ship code with agents.
How hidden instructions trick coding agents into unsafe work.
Read the researchEvergreen guideTool poisoning, overbroad permissions, and shadow MCP servers.
Read the researchNew dossierSlopsquatting is the supply-chain attack in which attackers register package names that AI models reliably hallucinate; when an agent writes an import for a package that never existed, the attacker’s registered code installs instead.
Read the researchNew dossierMalicious extensions and marketplace fraud place trojanized editor extensions, plugins, or CI actions inside trusted marketplaces, harvesting source code, prompts, and credentials from every developer who installs them.
Read the researchKeep exploring
The engineering desk continues across the ecosystem: security research on shipping agents safely, and plugins that put the standards to work.