VoicesKarpathy: Autoresearch 333 Experiments + AgentHub SETI@home VisionX/Twittervia @karpathy·high signalXBlueskyLinkedInCopy link630-line Python script, 35 agents ran 333 experiments overnight, ~20 improvements found. Building AgentHub for distributed agent research swarms.SourceSource pageX/Twitter↳ Follow the threadPolicy dependency / Stack layerClaude Code 2.1.259 refuses to start when a managed settings file cannot be parsedClaude Code ChangelogPolicy dependency / Stack layerDistilling 1,000 ML repos into 5,000 verified skills lifted an agent 134% on MLE-bench with the model and budget held fixedarXiv 2609.02749Policy dependency / Threat pattern"I Have Been Clawed" Published a CC BY 4.0 JSON Dataset of Coding-Agent Incidentsihavebeenclawed.com (Show HN)Policy dependency / Stack layerOpenAI tells House Democrats it is building 'automated shutdown capabilities' but withholds the hack logs Congress asked forReutersPolicy dependency / Stack layerCROSS-CATEGORY: Four Unrelated Vendors Shipped Agent Authorization Control Planes Inside 48 HoursJetStream (corroborated by Genesys Xperience 2026 coverage, aiagentstore.ai and Hacker News Show HN)Policy dependency / Stack layerFour advisories land on Databricks' Omnigent meta-harness, one critical, all reported by an autonomous security agentGitHub Security AdvisoriesStack layer / Update threadTip: one paragraph cuts Fable 5.1's unrequested fixes and surplus test files with no measured loss in task successAnthropicStack layer / Update threadThree Agent-Telephony Products Shipped on the Same Day at $0.02 a MinuteStartupCorners product digest (corroborated by Product Hunt and Show HN)