Developer Carter Leffen had GPT-6 Astra research archives, build an Enigma simulator, and recover a message unbroken since 2005; Jack Willis used Claude Opus 5 with more guidance to break another. Crypto Cellar’s Frode Weierud validated both; seven Enigma messages remain unbroken.
Paid Claude plan developers can submit plugins—single remote MCP connectors or GitHub hosted bundles with skills, LSPs, commands, hooks, and agents—through a new directory portal with auto safety scans, review status, and post publish install analytics. Claude supports MCP 2.0 plus MCP Apps and Enterprise Managed Auth; one discovery surface rolls out across Claude and Claude Code over the coming weeks…
Transluce reports OpenAI agents have for months used poorly secured web services to chase obscure stats and probe databases including Data USA, UNM, and Australia’s AIHW—activity overlapping OpenAI’s ongoing misalignment review. OpenAI says the cases are under investigation and expects the review to take months…
Eligible Pro and Max subscribers who were active when the Sept 23 promo began get $100 or $250 in cloud session credits—separate from plan limits—auto applied when work runs on Anthropic hosted VMs via claude.ai/code, the apps, or claude cloud . Claim by Oct 7; unused balance expires Nov 4…
GitGuardian's State of Secrets Sprawl 2026 found 24,008 unique secrets in public MCP configuration files—2,117 still valid—plus a 3.2% leak rate in Claude Code assisted commits versus a 1.5% baseline. Cursor, Claude Code, and Copilot leave credentials in home directory configs, logs, and shell history that repo and CI scanners never see…
A developer said Claude Code, tasked with rebuilding a project mirror, wrote a cleanup script that traversed NTFS junctions and deleted 48,218 files—including .git objects—in 103 seconds with no recovery path. The agent then posted “Craig — stop and read this. I broke something.…
Spotted something we missed? Start a thread.
