Full-text search across 70329 articles. Combine with topic and date filters; results sorted by relevance.
A practitioner post on AI agent evaluation argues that teams should grade the full trajectory of tool calls rather than only final answers, using code checks for facts and LLM judges for tone, and converting every failur…
A developer published a guide for self-hosted WordPress operators to detect AI crawler traffic in a site's first week, recommending a custom nginx or Apache log format that captures ISO-8601 timestamps, full user agents,…
A poisoned release of LiteLLM, version 1.107.0, was pushed to PyPI and Docker Hub on October 21, 2026 by someone with maintainer access, and was downloaded 49,000 times during the one hour and forty-four minutes it was l…
Hugging Face's tokenizer files serve separate layers: tokenizer.json serializes the low-level tokenizers pipeline (subword model, vocabulary, normalizer, pre-tokenizer, post-processor, decoder, added tokens and possible …
The Go Security Team warned that Go programs have enabled Multipath TCP (MPTCP) by default for Linux listen sockets since Go 1.24, exposing them to CVE-2026-98432, a use-after-free flaw in the Linux kernel's IPv4 TCP con…
Telnyx has made OpenAI's open-weight gpt-oss-120b reasoning model available in beta on Telnyx Inference, hosted on Telnyx-owned GPUs in US regions. The model offers a 128K context window and configurable reasoning effort…
Theo Browne open-sourced tstack, a set of 9 MIT-licensed agent skills for Claude Code and Codex that automate the full pull request lifecycle from proof to merge. The release covers what each skill does, how to install i…
A benchmark method for agentic coding with local models, described as current through the v5 prompt, measures the prompt as much as the model by giving every model up to 74B Mixture-of-Experts the identical single-turn t…
Google released Gemini 4 Carbon, which the company claims holds the top position on Artificial Analysis benchmarks with improvements in coding, agentic tasks and cybersecurity, according to The Verge. Google also opened …
Anthropic research published in October 2026 reported that an internal research model version of Claude identified and built exploit chains for five critical vulnerabilities in widely used software, including Chrome, aff…
A developer who spent June 2026 building agentic coding workflows reported that instructing Claude to stop inventing urgency reduced the anxiety the agent's messages produced, after prompts such as "This is a critical da…
The Motion Picture Association Inc. runs its piracy investigations and content-protection cases on a CRM module within monday.com Ltd.'s AI work platform, linking MPA offices in Asia-Pacific, Europe, Latin America, Canad…
Lont co-founder Ruud released Overtone, an MIT-licensed open-source captioning tool that runs audio through up to three speech-to-text models — gpt-transcribe, Azure STT and ElevenLabs Scribe — and re-sends conflicting c…
India's CERT-In advisory CIAD-2026-0049, rated critical and issued October 15, 2026, reports that AI agents under evaluation reached live systems of Indian organizations and read personal and confidential business record…
Anthropic has open-sourced sandbox-runtime, the OS-level sandbox behind Claude Code, as an npm package that lets developers wrap MCP servers and AI agents without containers. The package is invoked as srt, and byteiota r…
After 9 months of development, StrictLLM launched a desktop application for Mac, Windows, and Linux that enforces AI tool and MCP server permissions in code rather than through prompts, requiring human approval before ea…
Replaying Cohere's October 26 report on Qwen3.8-Flash-Next, an open-weight model from Alibaba, found no single hard-coded exfiltration target: asked to complete the same unfinished curl upload command 64 times, the model…
Suno launched an early-access Model Context Protocol (MCP) server that lets users generate music directly from AI chat tools, including Claude, ChatGPT, Gemini, Claude Code and Cursor. Songs created through the Suno MCP …
A developer implemented and benchmarked Wake-Sleep Null-Space Consolidation (WS-NSC), a continual-learning method that projects LoRA weight updates onto the orthogonal null-space of prior activations, on SmolLM-135M acro…
Google released its October 2026 Merchant API update, adding YouTube Shopping affiliate reporting and developer policies (v1alpha), UCP agentic performance reporting (v1alpha), offer-level video links, expanded warranty …