DeepSeek V4-Flash Adds Vision at No Price Premium
DeepSeek added image understanding to its cheapest fast model on August 21, 2026, billing images at standard V4-Flash rates with a free Files API.
Slack Code Puts AI Agents in Shared Team Channels
Slack shipped Slack Code on August 20, 2026, a channel type where a team and its AI coding agents work in the same shared thread.
Mistral Agentic Search Turns RAG Into a Search Loop
Mistral shipped Agentic Search on August 20, 2026, replacing one-shot document retrieval with a navigable loop the model drives itself using five file-like operations.
Claude's Agent Stack Goes GA: Browser, Skills, Files
Anthropic moved four agent building blocks to general availability on the Claude Platform on August 20, 2026: computer use, a new browser use tool, the Skills API, and the Files API.
DeepSeek Harness: Open Agent Framework, Everything a Plugin
DeepSeek Harness is an open, MIT-licensed framework for building AI agents where every component is a plugin. Here is what shipped and how to run it.
NVIDIA Nemotron 3.5 Lightning: Open 30B Agent Model
NVIDIA released Nemotron 3.5 Lightning, a 30B open-weights MoE model for always-on agents, alongside NeMo Switchyard, an open-source router that cuts agent cost by escalating only the hard steps to a frontier model.
Agent Plugins: OpenAI, Microsoft, Vercel Back One Standard
OpenAI, Microsoft, Amazon, Vercel and Cursor agreed on Agent Plugins, an open standard that packages MCP servers and Agent Skills into one portable folder that runs across AI tools.
AI Coding Agents: Humans Miss 1 in 3 Threats
A study of 40,000+ sessions found humans miss about one in three threats when approving AI coding agent commands. Here is why the human-in-the-loop leaks.
Google Cloud API Gateway Now Routes to Any LLM
Google shipped model routing for Cloud API Gateway on August 4, 2026: one OpenAI-compatible endpoint that forwards each request to Gemini, Claude, or an OpenAI GPT model.
Voice Coding Tools 2026: On-Device vs Cloud Compared
If you code under an NDA or on a private repo, an on-device voice coding tool is the only safe way to talk to your AI agent. SKI, Spokenly, and VoiceMode keep every word local. Here is how the leading voice coding tools compare.
Termexo: One Windows Workbench for Claude Code, Codex
Termexo is a free, local Windows workbench that runs Claude Code, Codex, and other coding agents in one grid of terminals, launched August 2, 2026.
QM: YC Open-Sources Its Multi-Agent Team Harness
Y Combinator open-sourced QM, an MIT-licensed multiplayer agent harness that gives every teammate an isolated workspace and drives Claude Code, Codex, OpenCode, and Pi from one core.
SKI: Free On-Device Voice Coding for Claude Code
SKI is a free desktop app that lets you talk to Claude Code, Cursor, Codex, and Gemini CLI out loud, with speech-to-text and voice both running fully on-device.
NVIDIA NOOA: Build AI Agents as Python Objects
NVIDIA open-sourced NOOA, a model-agnostic framework that turns an AI agent into an ordinary Python object. Here is how it works, the benchmarks, and how to build your first agent.
Gemini Spark Can Now Browse Chrome For You
Google Gemini Spark can now drive Chrome to complete multi-step web tasks using your logged-in accounts, and expanded to 160-plus more countries on July 30, 2026.
Perplexity Projects: A Multiplayer AI Workspace
Perplexity Projects turns Computer into a shared, multiplayer workspace for people and AI agents, with a persistent file system and Brain memory that carries context between tasks.
Gemini Agents Get Hooks, Budget Caps, Free Tier
Google shipped a major Gemini API Managed Agents update on July 28, 2026, adding environment hooks, a token budget cap, cron scheduling, an Environments API, and free-tier access. Gemini 3.6 Flash is now the default agent model.
MCP Goes Stateless: The 2026-07-28 Spec Explained
The Model Context Protocol 2026-07-28 spec drops stateful sessions for a stateless request/response model, so MCP servers can scale serverless. What changed, what breaks, and how to migrate.
Perplexity Personal Computer Lands on Windows
Perplexity has brought its Personal Computer AI agent to Windows, reaching more than a billion devices with local file and Microsoft 365 automation.
Buzz: Block's Open Workspace for Humans and Agents
Block released Buzz on July 21, 2026, a free open-source workspace where AI agents and people collaborate as equals, built on the decentralized Nostr protocol.
Devin Outposts: Run the AI Coding Agent on Your Hardware
Cognition shipped Devin Outposts on July 21, 2026, moving the coding agent's execution onto your own GPU, VM, Kubernetes cluster, or Mac mini while planning stays in the cloud.
Unabyss Gives AI Agents Portable Memory via MCP
Unabyss launched a Claude-focused, MCP-native context layer that carries your memory and preferences across Claude, Cursor, ChatGPT, and 50+ apps.
Amp Launches Puck, a Meta-Agent for Coding
Amp launched Puck, a meta-agent that acts as a home base for spawning and coordinating coding agents inside your harness.
ZooData Gives AI Agents a Structured Data Layer
ZooData, a new data layer for AI agents, launched July 18, 2026 and topped Product Hunt. It turns any URL into clean, agent-ready JSON via API, CLI, and MCP.