{"slug": "give-agents-the-bug-report-unprompted-they-fix-under-5-of-bugs", "title": "Give agents the bug report: unprompted, they fix under 5% of bugs", "summary": "SWE-sweep tested autonomous bug fixing by asking models to find and fix real bugs across 100 repositories with no ticket to guide them, and the best configuration fixed under 5% of bugs. The result indicates autonomous bug hunting still requires a human to specify what is wrong.", "body_md": "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.\nRead: 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.\nRead: Anthropic's Claude Code team introduced mods, TypeScript extensions shipped inside plugins that can alter behavior, customize the interface and add features.\nRead: 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.\nRead: 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.\nRead: DeepSeek released packaged desktop builds of DeepSeek Harness v0.2 preview for macOS and Windows. Linux users install it through the dsh npm package.\nRead: 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.\nRead: Ethan Mollick revised his earlier view that managing agents requires careful organizational design, arguing the Bitter Lesson now applies to agent scaffolding.", "url": "https://wpnews.pro/news/give-agents-the-bug-report-unprompted-they-fix-under-5-of-bugs", "canonical_source": "https://www.vibeleaderboard.ai/intel/brief/2026-10-02", "published_at": "2026-10-02 11:18:18+00:00", "updated_at": "2026-10-02 12:39:30.968296+00:00", "lang": "en", "topics": ["ai-agents", "ai-research", "artificial-intelligence"], "entities": ["SWE-sweep", "Anthropic", "Claude Code", "Earendil", "Pi 1.0", "Cloudflare", "DeepSeek", "Ethan Mollick"], "also_reported_by": [], "alternates": {"html": "https://wpnews.pro/news/give-agents-the-bug-report-unprompted-they-fix-under-5-of-bugs", "markdown": "https://wpnews.pro/news/give-agents-the-bug-report-unprompted-they-fix-under-5-of-bugs.md", "text": "https://wpnews.pro/news/give-agents-the-bug-report-unprompted-they-fix-under-5-of-bugs.txt", "jsonld": "https://wpnews.pro/news/give-agents-the-bug-report-unprompted-they-fix-under-5-of-bugs.jsonld"}}