Fast Mode Deadline, Postgres Delay

Claude Code 2.1.319 adds background MCP tasks and a deny-instead-of-prompting setting for headless runs, Cursor adds realtime voice chat with agents, and AWS ships persistent Kiro bots. We also cover Meta's Conductor acquisition and DigitalOcean's Cuttlefish iPhone app, and teach the difference between blocking and background tool calls and between agent memory and model retraining. Hosts: Alex & Jules. New episodes daily.

Apiiro reports thousands of secrets leaking from Claude Code's "Don't Ask Again" approvals file, and we cover what to rotate. Claude Code 2.1.319 adds background MCP tasks and deny-instead-of-prompt, and we teach how agent memory works, including Anthropic's "dreaming" research. Plus NVIDIA's Vera CPU for agents. Hosts: Alex & Jules. New episodes daily.

A LangChain study claims Stagehand beat Playwright MCP on cost, and we teach how caching lets an agent skip repeat model calls. Plus Phone Harness gives Claude Code a real phone, Stacktrace and CrowdStrike ship agent controls, and a Claude Code fast-mode deadline on November 30. Hosts: Alex & Jules. New episodes daily.

Cursor Apps turns a description into a hosted app with database and auth, Antigravity CLI plugs into any editor via ACP, and Codex CLI, Copilot CLI, Amp and Claude Code all ship updates. We also teach how the Agent Client Protocol (ACP) lets editors and coding agents talk over JSON-RPC.Hosts: Alex & Jules. New episodes daily.

Claude Code 2.1.317 adds a prompt before your account email goes out over the network, and we explain why anything in a model's context can leak into its outputs. We also cover Stroq, an open source action firewall for coding agents (and the idea of "tainted" sessions), Tencent's Hy4 model, and how hybrid attention layers cut long-context memory costs.Hosts: Alex & Jules. New episodes daily.

Anthropic's FlashProve prototype aims to cryptographically prove which model produced a response, Moonshot releases Kimi K3.1 Pro weights (2.8T parameters, 104B active), and we cover MCP OAuth redirect risks, Gemini's chat-training settings, and an RTX Spark Linux slowdown. Today we teach mixture-of-experts and zero-knowledge proofs in plain terms. Hosts: Alex & Jules. New episodes daily.

A GitHub issue alleges Postman auto-installed plugins and startup hooks into Codex and Claude Code. We also cover a reported Codex training-setting flip, the pushback on yesterday's weight-editing story, and Simon Willison running a 27B model locally. We teach quantization: how rounding a model's numbers makes it fit on a laptop. Hosts: Alex & Jules. New episodes daily.

Mistral Large 4 hits Hugging Face as a trillion-parameter open-weight model, and we explain how mixture-of-experts lets it run on a fraction of its parameters per token. Also: DeepSeek V4.1 Flash, a weight-editing experiment on Qwen, a pretraining-poisoning backdoor paper for agent builders, and Momentic's MCP-connected AI QA tool. Hosts: Alex & Jules. New episodes daily.

Claude Code 2.1.314 and 2.1.316 land, with Opus 4.8 fast mode set for removal on November 30. We also cover why Postgres 19 is late (LLM-found bugs), a 12-framework benchmark of what Claude Code costs per stack, and quick hits on Codex, Gemini CLI and OmniDiff. The episode teaches prompt caching and why switching modes mid-session costs more.Hosts: Alex & Jules. New episodes daily.

Cisco Talos details VOLTA, an infostealer that sends stolen data to an LLM to rank victims, and a researcher's chained-bug ChatGPT account takeover via a shared link. We cover Claude Code 2.1.311 and 2.1.312, plus Prospero, a lean C++ coding harness. Concepts taught: temperature, postMessage, and what an agent harness is. Hosts: Alex & Jules. New episodes daily.