Full-text search across 62885 articles. Combine with topic and date filters; results sorted by relevance.
An AI security researcher confirmed an unauthenticated remote code execution vulnerability in OpenMed, demonstrating how AI can move vulnerability research from scattered code signals to a reproducible attack path, SD Ti…
Pipper launched as a single Electron desktop interface for controlling multiple AI coding agents, built on the ACP protocol to support Claude Code, Codex, Cursor, OpenCode, and Grok out of the box. The app is available a…
Independent researcher and MCA graduate MadB0i is seeking an arXiv cs.CR endorsement for a first paper, "KavachBench: An Empirical Evaluation of Deterministic Policy Enforcement Against Issue-Based Prompt Injection in AI…
A developer building an autonomous coding system with Claude Code split work between a planner agent and fresh-context implementer sub-agents, but found that roughly one in three implementer runs solved the wrong problem…
The Ninth Circuit Court of Appeals upheld a lower court ruling against a group of open-source software developers who claimed GitHub and OpenAI violated the Digital Millennium Copyright Act by removing copyright manageme…
A developer running a headless Claude Code agent loop lost USD 2.79 and completed zero tickets on September 12 after two parallel coordinators reported they were "waiting" for background tasks but had actually terminated…
Vercel's AI SDK harness layer now supports GitHub Copilot through the official @ai-sdk/harness-github-copilot adapter, letting applications run different coding agents through the same HarnessAgent interface without chan…
A developer released codsh 0.22.1, a DeepSeek-native terminal coding agent that turns a one-sentence request via its /ship command into verified code. The update fixes grill cards so long questions wrap fully instead of …
Våld Labs has released Tresse, a dual-layer synthesizer plugin for macOS and Windows with 39 macro-based synthesis engines, 24 based on Mutable Instruments Plaits code and 15 newly developed, plus a generative sequencer.…
A well-targeted AI pause focused narrowly on recursive self-improvement (RSI) and related capabilities could increase the rate of non-AGI technological advancement in the short run, according to an analysis of AI race ju…
A blogger announced they will stop using AI to write their blog posts, citing an arXiv study (2412.07200) on AI-assisted essay writing that found revising AI-generated text improved writing quality while using it unchang…
A Hacker News post argues that the programmer using AI is the most important factor in AI-assisted coding, likening the programmer to the harness that determines output quality. The author contends that vibe coding revea…
Researchers introduced the Relational Hypergraph Transformer (RHT), a unified architecture for multi-table learning that represents relational databases as hypergraphs and uses pentadimensional embeddings (PentE) with sp…
A developer and their teammate built a tool for DataHub's 'Build with DataHub: The Agent Hackathon' that generates Airflow DAGs from DataHub metadata. The tool walks lineage graphs, reads tags and glossary terms, and map…
Jean, an open-source development environment for AI-assisted coding, is available under the Apache 2.0 license, free forever with no paid tiers or feature gates. It enables running multiple AI agents in parallel, each in…
A developer found that running the Go tool `deadcode` on their monorepo uncovered 2,300 lines of unused code, much of it generated by AI models following prompt patterns like 'Add retries with exponential backoff.' After…
A developer shares a workflow using Matt Pocock's AI coding skills to improve agentic coding quality. The skills, including /grill-with-docs, /to-spec, /to-tickets, /implement, and /handoff, help developers plan, specify…
SpaceXAI launched Grok Bot, a system of persistent AI agents that operate on dedicated cloud computers and can execute multi-step tasks across business applications, currently in beta for SuperGrok Heavy, Cursor Ultra, a…
Between October 2018 and July 2026, AI models progressed from simple systems like BERT to massive agents that solve complex math and write software, with the ability to resolve real coding issues improving by nearly six …
An AI agent session can consume 50,000–200,000 tokens, compared to a few hundred to a couple thousand for a chat turn, with multi-agent jobs exceeding one million tokens, according to illustrative ranges from practitione…