Full-text search across 26433 articles. Combine with topic and date filters; results sorted by relevance.
A developer built keymap, a dependency-free local Python tool that scans the config files agent tools, CLIs and shells read — including ~/.claude.json, ~/.codex/config.toml, ~/.env, shell rc files and history, and ~/.aws…
An independent benchmark run by Leanzero found that Mistral Large 4 scored 0.5462 on its Gauntlet agentic coding benchmark, ranking 28th of 44, and 0.1086 and 0.3304 on its two Forge 2.0 builds, trailing DeepSeek V4.1 Fl…
Mulga is exploring the formal specification language TLA+ as a way to preserve shared engineering understanding as LLMs take over code writing, using a real AWS EBS spec as an example. The company argues that since AI-wr…
The October 2026 SlopCodeBench update found that enabling XHigh reasoning effort on Claude Opus 5.5 costs 4.22x more while producing fewer correct results than Medium, with Opus 5.5 falling from 28 to 25 strict checkpoin…
Computer scientists at Binghamton University demonstrated a backdoor attack called MuTRAP that hides trigger words in large language models used to plan robot tasks, causing a Segway-based mobile robot to redirect its kn…
Zscaler ThreatLabz reported in October 2026 that it traced an AI-assisted hacking campaign combining a customized ARTEX penetration-testing deployment with Claude Code, Codex and Grok workspaces, extending well beyond th…
A developer working on the BibleDept project changed its waveform rendering so that live PCM sampling is only requested by the visible full player, while hidden or paused players skip the sampling listener and the mini-p…
A matched study of more than 100,000 GitHub accounts found that developers who moved from autocomplete to autonomous coding agents produced 180 percent more commits but only 30 percent more releases, with no measured inc…
A developer testing llama.cpp build b11753 found that llama-server's OpenAI-compatible /v1/responses endpoint reports truncated replies and tool calls as "completed" with incomplete_details set to null, while the same re…
A developer built GrammarBar, a free, open-source macOS menu bar app that fixes, polishes and translates text in place without leaving the current application. The app shells out to Claude Code's headless mode (defaultin…
An MIT CSAIL paper, "AI Disruption and Falling Costs in Theory Research," found that the share of cutting-edge theory problems where AI models matched human authors' conclusions rose from 3% with o1 to 35% with GPT-5.5 a…
A developer released llms.txt Auditor build 0.3.2, an Apify-hosted tool that checks a submitted public origin for llms.txt, llms-full.txt, and root robots.txt and reports observed policy evidence and explicit gaps rather…
A developer published a headless template for running Claude Code unattended, recommending that every reachable tool call resolve to an explicit allow or deny rule in settings.json rather than relying on prompts. The gui…
Meta paused a Muse feature that lets users share AI conversations via links after Claroc researchers found more than 6,800 Muse share links through public DNS records, exposing chats containing a marketing company's reve…
A developer built Refmade, a desktop app that visualizes Claude Code and Codex sessions as an office with a team lead and subagents, by coordinating five parallel AI builder agents from a single markdown contract (office…
Inception, a Redwood City startup, launched Mercury 3 on October 19, 2026, a diffusion language model that generates tokens in parallel and claims 1,068 tokens per second with a 1-million-token context window, positionin…
ElevenLabs released Eleven v4 and Eleven v4 Turbo on September 28, pitching the models at developers building voice agents with a median inference latency of about 100 milliseconds and median time to first speech near 15…
AWS released AWS Agent Registry, a managed service for cataloging and discovering Model Context Protocol (MCP) servers, agents and agent skills, so orchestrators can find tools at runtime instead of hardcoding them. Publ…
AI agent reliability depends less on the underlying large language model than on the surrounding "harness" of rules, tools, workflows, memory and guardrails, according to an analysis that cites Cole Medin's video using L…
Tabularis released v0.29.0, a release of 115 commits and more than 100 merged pull requests from 34 external contributors, adding ChatGPT as an AI provider so eligible ChatGPT Plus or Pro accounts can power SQL generatio…