Fetching from the wire…
01
02
03
04
05
06
07
08
Tools2026-08-15 · source-backed
NanoNets' Graft builds a persistent graph of linked markdown nodes describing subsystems, APIs and concepts, then injects ranked context per prompt through hooks instead of letting the agent re-explore the repo. A 162-run sweep across two repos reports 8,070 to 4,650 tokens, 4.2 to 2.3 tool calls, 39.8s to 15.8s latency, and $0.0429 to $0.0292 per task. On 50 SWE-bench Verified instances it resolved 33 versus 27 cold. GitHub 2.7k stars. graft init also adds a freshness statusline and post-edit blast-radius warnings.
Each link below shares sources, entities, or timing with this story.
This is the paper of the week. arXiv 2607.28871 introduces BSG-VA, which replays every validation command an agent runs across three code states: the original buggy code (B), the candidate patch (S), and the gold developer fix (G). If a test passes in all three states, it neve...
Alibaba released Qwen3.6-27B on April 22. Dense architecture. Open weights. 77.2% on SWE-bench Verified, within 3.7 points of Claude Opus 4.6. On SkillsBench, it scores 48.2% versus its own 397B MoE predecessor's 30.0%. That's a 77% improvement with 14.8x fewer parameters. Let...
I check Product Hunt maybe once a week and usually regret it. Today's board is worth reading as market structure. The July 30 leaderboard: SKI at 277 upvotes (free voice input for Claude Code and Codex). AI Search Console at 249 (prompt analytics and citation mapping). Memmy A...
A paper from Xiao Yu, Baolin Peng, and Ruize Xu makes a claim that seems obvious once stated and is genuinely new as a training methodology: modern agents are inseparable from their inference harnesses, so training them in stripped-down RL sandboxes produces a train/serve mism...
119 repository-level tasks from 98 GitHub repos across 20 scientific domains, split into issue-driven, expert-exploratory, and engineering-integration paradigms. Claude Code with Opus-5 (max) lands below 50%. arXiv The ablation is the better finding: stripping explicit scienti...
Egonex-AI/Understand-Anything converts a repo into a searchable, queryable knowledge graph with integrations for Claude Code, Codex, Cursor, and Copilot, under the banner "graphs that teach over graphs that impress." (GitHub) It's the second high-traction code-to-knowledge-gra...
MindPattern daily
One email a day at 7 AM. Sources and a take on every story. Unsubscribe anytime.