Fetching from the wire…
01
02
03
04
05
06
07
08
Vibe Coding2026-08-11 · source-backed
SkillSentry (arXiv 2608.09253) targets the gap where an agent completes a task under skill guidance then fails the same task on a repeat run. It defines a DSL for runtime guidance, initializes it from skill specs plus insights mined from historical successful and failed traces, wraps the execution loop to monitor and steer in real time, then refines from new traces. Across 15 skills on Haiku 4.5, Opus 4.6, GPT-5.2, and GPT-5.4: 24.1% average success lift, and, the underrated result, reduced variance across repeated executions. Variance is what makes a skill unusable in an unattended pipeline.
Each link below shares sources, entities, or timing with this story.
$3,054 against $38,370. Same benchmark, better score. Praxist (arXiv 2608.25955, submitted August 26) replaces per-attempt agent memory with a typed evidence graph of findings, plus lane-structured frontiers and agendas, so later attempts inherit validated mechanisms rather th...
Two thirds. Not two thirds of a contrived jailbreak set. Two thirds of realistic malicious issue requests, against the exact three tools most of the people reading this run daily. Ankur Singh, Jinqiu Yang, and Tse-Hsun Chen built IssueTrojanBench across four attack categories...
Terminal-Bench 2.1 results (entries dated June 17) put Codex CLI on GPT-5.5 first at 83.4%, Claude Code on Fable 5 second at 83.1%, and Claude Code on Opus 4.8 at 78.9%. The asterisk matters more than the ranking: Fable 5 and Mythos 5 have been export-suspended since June 12,...
SoL-Pi scales automated research loops over harness designs rather than over models, keeping four mechanisms that survived selection: action execution, context compaction, observation handling and delegated reading. On the 51-task EdgeBench evaluation with GPT-5.6 Sol and Opus...
Thibault Sottiaux at OpenAI published an investigation into "a handful of reports where GPT-5.6 unexpectedly deleted files," finding it happens most commonly when full access mode is enabled in Codex. Simon Willison relayed it. A frontier lab publishing a first-party post-mort...
July 9, across VS Code, Visual Studio, Copilot CLI, the cloud agent, github.com, GitHub Mobile, JetBrains, Xcode, and Eclipse. Sol is the high-reasoning tier at $5/1M in, $30/1M out, gated to Pro+/Max/Business/Enterprise. Terra is the balanced default at $2.50/$15. Luna is fas...
MindPattern daily
One email a day at 7 AM. Sources and a take on every story. Unsubscribe anytime.