Fetching from the wire…
Security2026-09-03 · source-backed
Stanislav Fort published on September 2 that AISLE's autonomous system produced 29 reports against curl, after curl founder Daniel Stenberg had posted publicly on August 24 that Mythos "can't find any more" and Codex security "shows an empty list." Six were accepted by curl's security team and assigned CVEs in 8.22.0, covering an OpenSSL provider use-after-free, a pinning bypass, CA-store connection reuse, a tab-based secure-attribute bypass, a wolfSSL CA-cache callback override, and a domain-scoped public-suffix cookie bug. All six rated Low. The count matters less than the setup: the zero-result baseline was public and timestamped before AISLE ran, and curl's maintainers decided what counted. That's a cleaner methodology than most AI security claims get.
Each link below shares sources, entities, or timing with this story.
OpenAI Devs announced on August 26 that WebMCP works in the ChatGPT desktop app's built-in browser and in ChatGPT Sites, so ChatGPT and Codex can call a site's declared tools directly. WebMCP is an experimental web standard adding navigator.modelContext to the browser, letting...
NVD posted nine advisories on August 25, clustering into one shape: a local server assuming a browser can't reach it. PraisonAI validated MCP origins with request_origin.startswith(allowed) against a localhost allowlist, so an attacker-registered localhost.attacker.com passes...
PR #1218 in a Snowflake repository replaced a safe pattern with an unsafe one. The old code used env: plus jq --arg to pass an issue title into a shell step. The new code interpolated github.event.issue.title directly into a run: block. That's the textbook GitHub Actions scrip...
A community hackathon run July 15 to August 2 had 1,221 participants use Claude Code, Codex and Cursor to reproduce 2,226 of ICML 2026's 6,352 accepted papers, producing 6,816 logbooks and 2,962 cloud jobs (Hugging Face). 51% of examined papers had at least one claim verified,...
Anthropic released Claude Fable 5.1 on September 1. Claude Code v2.1.257 made it the default Fable model at 17:53 UTC that day, with a 1M-token context window, $10 per million input tokens, $50 per million output, and $0.25 per million on cache reads (claude-code CHANGELOG). B...
Simon Willison spent a while taking ChatGPT Work apart and published the map on August 30. Work splits into Work Cloud and Work Local, the latter being the renamed Codex desktop app, at $20/month and up since July 9. He enumerates six capabilities Work has that Chat doesn't, a...
MindPattern daily
One email a day at 7 AM. Sources and a take on every story. Unsubscribe anytime.