The Brief

New briefs daily around 7 AM Eastern

Give agents the bug report: unprompted, they fix under 5% of bugs

SWE-sweep asked models to find and fix real bugs across 100 repositories with no ticket to guide them. The best configuration fixed under 5%, so autonomous bug hunting still needs a human to say what is wrong.

• 01Read

  Give agents the bug report: unprompted, they fix under 5% of bugs

  SWE-sweep asked models to find and fix real bugs across 100 repositories with no ticket to guide them. The best configuration fixed under 5%, so autonomous bug hunting still needs a human to say what is wrong.

  lieret→
• 02Read

  Claude Code adds mods, TypeScript plugins that change behavior and UI

  Anthropic's Claude Code team introduced mods, TypeScript extensions shipped inside plugins that can alter behavior, customize the interface and add features.

  ClaudeDevs→
• 03Read

  Earendil ships Pi 1.0 harness and Pi Durable for crash-safe agents

  Earendil released Pi 1.0, a minimal agent harness that adds Codemode, deferred tool loading and Anthropic cache warming, alongside Pi Durable, an experimental base for long-running agents that checkpoints state.

  sergiotapia→paulsmith→[AINews] Pi 1.0, Pi Durable, and AIE NYC→
• 04Read

  Cloudflare open-sources Clef decision models with RL fine-tuning

  Cloudflare released Clef and Clef-flash, open-weight models for classification and routing with bounded structured outputs, hosted on Workers AI, plus a platform to fine-tune them with reinforcement learning.

  Introducing Clef: our open-source decision models, and new RL fine-tuning platform→Cloudflare→
• 05Read

  DeepSeek ships Harness v0.2 preview as macOS and Windows apps

  DeepSeek released packaged desktop builds of DeepSeek Harness v0.2 preview for macOS and Windows. Linux users install it through the dsh npm package.

  deepseek_ai→easonlu1017→
• 06Read

  Paper finds 47 vacuous completion gates across 69 shipped agent loops

  A new arXiv preprint defines bounded agent loops as a worker, a gate it cannot write to, and a declared budget, and audits shipped loops for gates that pass without checking anything.

  vpbhardwaj→
• 07Read

  Ethan Mollick revises his case for hand-built agent orchestration

  Ethan Mollick revised his earlier view that managing agents requires careful organizational design, arguing the Bitter Lesson now applies to agent scaffolding.

  Ethan Mollick→