Meta and Shopify pair Muse with Shop Pay checkout
Zuckerberg announced a Shopify team-up to make shopping and checkout easier in Muse. Shopify plans the integration with Shop Pay; operational details remain unknown.
News
SpaceXAI ships Grok 4.7 for coding and knowledge work
SpaceXAI’s Grok 4.7 lands in Cursor, Grok Build, and the API at Grok 4.6 prices, with longer-task training and a native Grok Bot harness.
Amazon blocks Meta’s Muse from agentic shopping on amazon.com
Amazon cut off Meta’s Muse shopping agent, citing unauthorized access, stored logins, and Conditions of Use after asking Meta to opt Amazon out.
Berkeley HarnessTax: coding harnesses move cost more than success
UC Berkeley’s HarnessTax finds Claude Code, Codex CLI, and Pi change coding-agent success little while token cost can rise about 5×.
MCP Over CLIs, Costly Agent Failures, and Why Compaction Isn’t a Filter
13–19 Sep 2026
SpaceXAI ships Grok Voice Transcribe 2.0 at the same STT price
SpaceXAI released Grok Voice Transcribe 2.0 for batch and streaming STT at $0.10/$0.20 per hour; docs already list 2.0 as the Speech-to-Text API default.
MIT and Sakana propose SIFT for cheaper agent self-improvement
MIT and Sakana’s SIFT paper uses LLM judges and tree search so coding agents self-modify with far less full-benchmark evaluation.
Google Labs expands CC into a shared household agent
Google Labs turns CC into a shared household agent with its own Google account for up to six U.S. adults, plus a daily brief and waitlist.
Claude Code 2.1.278 defaults auto mode to a server-side classifier
Claude Code 2.1.278 defaults auto-mode checks to a server-side classifier for API, Enterprise, and cloud paths, with no classifier overhead when that path works.
Playwright: agents can pick MCP or CLI
Playwright says pick MCP or CLI; docs spell out best-for, token cost, headed vs headless, and how to install.
Claude Code 2.1.277 falls back to AGENTS.md
Claude Code 2.1.277 reads AGENTS.md when CLAUDE.md is missing, with a /config toggle, plus gateway proxy headers and hang-exit fixes.
Claude Code Projects now run parallel cloud threads from one chat
Anthropic turned Claude Code Projects into one conversation that coordinates parallel cloud threads and keeps working after you close the laptop.
Google Home MCP lets Claude and OpenClaw control Home devices
Google opened Home MCP to Claude, OpenClaw, and other agents for U.S. Premium Advanced homes; door unlocks are blocked and Google warns the wrong agent can still misbehave.
Anthropic publishes three metrics on AI-led R&D, agent oversight, and compute
Claude leads 26% of AI R&D as of August; ~30k agents are monitored; a July 13–20 week put about 6% of R&D compute on safety.
Claude Code 2.1.276 fixes gateway 400s on advisor_20260301
Claude Code 2.1.276 restores proxy and gateway traffic after 2.1.275 400s on Input tag advisor_20260301.
Claude Code 2.1.275 adds send-now and account skill sync
Claude Code 2.1.275 adds send-now to flush queued mid-turn messages and syncs claude.ai skills/plugins; it also fixes a terminal API 400 behind rewriting gateways.
TensorRT Edge-LLM posts a 6.4× MLPerf Edge Agentic run on Jetson AGX Thor
NVIDIA says TensorRT Edge-LLM finished MLPerf Edge Agentic 6.4× faster than llama.cpp on Jetson AGX Thor, at 52.33 tok/s on Qwen3.6-27B.
OpenAI ships Astra for Law as a GPT-6 Astra legal configuration
OpenAI launches Astra for Law: GPT-6 Astra plus a U.S. legal search index, with Trusted Access in ChatGPT and Codex before the API.
OpenAI shows how ChatGPT Work and Codex analytics map usage to outcomes
OpenAI’s Admin Console walkthrough shows how ChatGPT Work and Codex credits map to classified tasks and merged-code outcomes so admins can argue value, not just spend.
Composio: GPT-6 Astra success is close across six harnesses; time and failed-run tokens are not
Composio’s Astra bench finds a ~7-point success spread (72.4%–65.5%), 122s–194s per task, and 3–5× tokens when runs fail.
OpenAI publishes a framework for reporting model misalignment
OpenAI set disclosure criteria and process deadlines for disclosing model misalignment and released six reports of unexpected agent and model behavior.
Anthropic merges Claude chat and Cowork into one assistant
Anthropic is folding Claude chat and Cowork into one assistant, adding Docs and Slides, and rolling the change out first to Pro and Max.
Claude Code 2.1.274 bounds MCP startup waits and warns on critical memory
Claude Code 2.1.274 adds a critical-memory warning, CLAUDE_CODE_MCP_STARTUP_WAIT_MS, OpenTelemetry events, and a large MCP and VS Code fix set.
OpenAI tests Sponsored Agents in ChatGPT ads
OpenAI is testing labeled, advertiser-backed Sponsored Agents after ChatGPT ads, plus Ads Manager prompts and HubSpot and Shopify integrations.
Command Code ships a beta desktop app for Mac, Linux, and Windows
Command Code’s beta desktop app for Mac, Linux, and Windows bundles the agent runtime with files, Git, terminal, previews, and /design.