Newly Accepted

Google’s announcement for new Gemini models focused on agentic workflows, coding, multimodal work, and cybersecurity use cases. The page includes performance claims, pricing, safety notes, and example applications.

A 30-second WebGPU benchmark that measures a device’s memory-bandwidth ceiling for local AI and checks for a WebGPU subgroup correctness bug, with receipts, projections, and comparison data.

CodeTrain is a hands-on AI coding trainer that teaches developers on their own codebase by making them write each step themselves, with hints, test runs, and progress tracking.

An MCP server for GoCD that lets AI agents inspect and operate pipelines using per-user GoCD personal access tokens. It exposes read, action, and destructive toolsets, plus resource URIs for dashboard, agents, and pipeline data.

A Go-based CLI for managing Python virtual environments, with MCP tools for AI agents, typed JSON snapshots, ephemeral execution, and file-watcher sync for evolving scripts.

A local-first CLI that compresses application logs into a token-optimized Markdown/JSONL summary for LLMs.

Custodian Labs is a developer platform for building and deploying AI agents from a small code snippet. The page highlights agent deployment, built-in RAG, model flexibility, multi-agent routing, and a privacy-focused Guardian layer for handling PII.

A macOS Model Context Protocol server that gives AI agents local voice output and persona-based speaking styles.

A blog post describing Annex, a local-AI security tool that encrypts selected files during agent sessions to keep sensitive data away from local AI tools. The post explains the threat model, encryption flow, and the stack used to build it.

An open-source protocol ecosystem for controlling LLM behavior during software development, with a focus on preventing context drift and code regressions through minimal diffs, inspection-first workflows, and snapshot-based session handoff.

VS Code extension that helps review large AI-generated diffs with local call graphs, file dependency graphs, and review cards that show blast radius and context before changes are pushed.

A GitHub repository for a self-correcting algebraic agent swarm runtime. The README frames it as a bounded, cybernetic system for multi-agent workflows, with separate components for intake validation, intent bridging, and swarm control.

Self-hosted Hacker News reader that summarizes top stories with AI, translates them into 20+ languages, personalizes the feed, and sends a daily briefing to Telegram.

Open-source personal AI companion that runs on your own machine, with persistent memory, background tasks, local voice, browser automation, and support for MCP and common personal data sources.

A curated dataset of sandboxes and related approaches for running AI coding agents with isolation, network control, or credential mediation.

A web-based multi-model comparison of procedural medieval castle worlds made with three.js. It showcases the same castle brief across eight models and three art directions, with static gallery previews and full-page worlds you can open.

A developer profile platform that pulls in contributions from external sources and adds an embedded AI hypeman for answering questions about your profile. It also offers developer APIs for using the profile data in other apps.

A Claude Code skill that searches royalty-free stock photos through the Pexels API, with install and usage instructions in the repo.

A private AI meeting-notes app that transcribes calls locally and uses Codex, Claude Code, or a local LLM to generate summaries and action items on-device.

A local-first AI reading assistant for Chrome that rewrites pages into customizable briefs, supports reusable reading recipes, and can run with local or hosted models.

Local codebase indexer and MCP server for Claude Code that watches project files, builds a private index on-device, and serves relevant snippets to Claude and other MCP clients.

AI-powered mock interview tool that generates role-specific questions from your CV and job description, scores answers, and gives concrete feedback and rewrites.

AI Buddy is a free macOS menu-bar app for screenshots and voice dictation. It lets users capture regions, windows, or full screen, copy instantly to clipboard, and transcribe speech to text using Gemini 3.5 Flash.

Google Research paper describing a deployed multimodal defense system for detecting coordinated synthetic spam and abusive media at scale, using account relatedness, synthetic content classification, and LoRA-tuned LLMs.