AI Tools Daily

Moonshot AI drops Kimi K3 — a 2.8-trillion-parameter open MoE with 1M context and weights coming July 27 — hours before Gemini 3.5 Pro misses its launch window. Plus Claude Code 2.1.212's runaway-agent caps and NVIDIA's open Nemotron 3 Embed models, with plain-language lessons on why agent loops need budgets and what embeddings actually are. Hosts: Alex & Jules. New episodes daily.

Claude Code 2.1.211 ships subagent text streaming and a quietly huge performance fix — up to 7x faster tool rounds when you run lots of MCP tools. Then the big one: the next Model Context Protocol spec (final July 28) drops sessions entirely and goes stateless, so remote MCP servers scale like any boring web API. We teach stateful-vs-stateless and why loading too many MCP tools silently taxes your agent, plus open-weight news (Inkling, GLM-5.2, MiniMax M3). Hosts: Alex & Jules. New episodes daily.

Claude Code v2.1.210 is a security-first release: the agent is hardened against indirect prompt injection, a sub-agent worktree-isolation hole is patched, and the permission classifier now defaults to Sonnet 5. We teach what indirect prompt injection actually is and why privilege separation shrinks the blast radius — plus a look at Grok Build's agent ergonomics polish. Hosts: Alex & Jules. New episodes daily.

Claude Code's desktop app adds a built-in sandboxed browser — and ships the safety rails in the same release. Plus Cursor 3.11 (side chats + an iPhone app), Cognition's SWE-1.7 at 1,000 tokens/sec, Grok 4.5's self-disclosed benchmark contamination, and free-tier background agents in the Gemini API. We break down what "sandboxed" actually protects and why you should distrust a leaderboard number. Hosts: Alex & Jules. New episodes daily.

Claude Code adds a fallbackModel setting that keeps long sessions alive when your primary model is overloaded, X launches hosted MCP servers for Cursor and Claude, and AgentPrizm ships a governed persistent-memory layer for agents. We teach why models get "overloaded," the difference between local and hosted MCP servers, and why language models forget everything between sessions. Hosts: Alex & Jules. New episodes daily.

Claude Code 2.1.207 makes agent auto mode the default on enterprise backends and squashes the terminal-freeze bug, Cursor 3.11 adds side chats to keep your main agent focused, and Windsurf lets subagents call MCP tools directly. Plus a leaderboard check on open weights — and a plain-English breakdown of Mixture-of-Experts and why "active parameters" is the number that actually predicts your cost. Hosts: Alex & Jules. New episodes daily.

Claude Code 2.1.206's /doctor now offers to TRIM your CLAUDE.md — cutting instructions the model can already derive from the code — and we explain why a leaner context beats a bloated one. Aider hits 93% on its edit benchmark with a rewritten "repo map," so we teach how a coding agent decides which slices of a huge codebase to actually show the model (retrieval, the cousin of RAG). Plus OpenCode passes 165K stars and rebrands to Anomaly, and Windsurf's Devin Local adds ACU usage display and faster MCP startup. Hosts: Alex & Jules. New episodes daily.

GPT-5.6 launches publicly today as three tiers — Sol, Terra, and Luna — across ChatGPT, Codex, and the API. We break down Kimi K2.7 Code landing in GitHub Copilot on a US host and use its pricing to teach prompt caching, then cover MCP's biggest-ever rewrite going stateless (and what stateful vs stateless actually means for your infra costs). Plus a Claude Code headless fix and why model routing is now the default. Hosts: Alex & Jules. New episodes daily.

Mistral 3 ships open-weight — small dense models plus a "fat but sparse" 675B mixture-of-experts — so we finally explain what MoE actually means and why active-vs-total parameters is the whole story. Plus Microsoft wires Dataverse into Claude, Cursor, and Copilot over MCP with real governance (the defense against yesterday's agentjacking), and Claude Code's background agents now open their own draft pull requests from isolated worktrees. Hosts: Alex & Jules. New episodes daily.

We break down "agentjacking" — a new attack where a fake Sentry bug report tricks Claude Code or Cursor into running malicious commands via MCP — and teach the prompt-injection flaw behind it: why a model reads instructions and data in the same breath. Plus the Fable 5 inclusion window closing today, and GitHub Copilot's CLI going model-agnostic with Kimi K2.7 alongside Claude. Hosts: Alex & Jules. New episodes daily.