Researchers using frontier AI models turned two decade-old PHP vulnerabilities, dubbed PHP2Shell and rated CVSS 9.2, into working unauthenticated remote code execution exploits within weeks, prompting…
A developer outlined a scheme for letting AI agents make payments using "agent cards" — virtual cards scoped to a single agent and a single job, with spending rules enforced on the card rather than in…
A developer extended the open-source LINE AI Business Card Assistant (line-card-assistant) with a Cloud SQL PostgreSQL-backed PostgresStore, completing a three-part cloud-native persistence scheme alo…
A practical checklist for agencies advises vetting AI tools on five questions before uploading client documents: where text goes, how long it is kept, whether it trains a model, who else processes it,…
A developer built Sloth Errands, an open-source overnight agent that completes routine web chores such as renewing library books, booking gym slots and checking parcels using OpenAI's open-weight gpt-…
A developer argues that AI assistant memory should live in plain files rather than inside a single product's account, so the same directory can be shared across ChatGPT, Claude, Cursor, Gemini CLI and…
A developer has documented a repeatable map-to-model pipeline that pairs CADMapper, a map-to-CAD converter drawing on OpenStreetMap, NASA and USGS data, with GenAnywhere, an AI 3D scene generator, tre…
A developer open-sourced Slipmark, an Apache-2.0 verification gate that records the SHA-256 hash of every file an AI agent's verdict is based on and re-hashes those files before merge, refusing any ve…
Ginkgo Bioworks connected GPT-5 to its cloud laboratory in February 2026, running 36,000 experimental conditions across six iterative cycles in which the AI designed, executed and learned from physica…
A developer built a production RAG document Q&A chatbot on Amazon Bedrock using NestJS, PostgreSQL with pgvector, hybrid search with reciprocal rank fusion, prompt caching, and Claude, and documented …
A developer rebuilt QuotaTray, a macOS and Windows menu-bar utility that displays AI API quota usage, as a native app after an initial Electron version, shrinking the macOS Apple Silicon build from 11…
A developer built Sloth Organizer, an open-source, local-first file organization app that uses a locally hosted model via LM Studio to scan messy folders and propose organization plans, then moves or …
A developer released soma-protocol, an MIT-licensed MCP endpoint that combines agent memory, signed obligations and offline-verifiable proofs in one service. The system stores facts outside the contex…
A comparison of major AI governance frameworks—NIST, ISO 42001, the EU AI Act and others—aims to help teams choose among them when building responsible AI solutions. The piece frames compliance naviga…
A developer published a checklist for shipping single-file HTML artifacts generated by LLMs (Claude, ChatGPT, Cursor) to non-technical stakeholders, covering seven pitfalls: broken relative asset refe…
A developer published a seven-step guide for protecting production databases from AI coding agents, covering controlled backups, restore verification, a separate development database, and settings tha…
Security firm Calif, working with OpenAI's GPT-5.6 Sol model through OpenAI's Patch the Planet program, discovered two critical PHP vulnerabilities (CVE-2026-109618 and CVE-2026-109619, both rated 9.2…
A developer self-hosted Ai2's AstaBrief-8B, an Apache 2.0 cited-report model built on Qwen3-8B, on a single 24 GB RTX A5000 and measured its inference behavior and citation accuracy. Across 11 generat…
Anand Biju built logsloth, a command-line tool that parses long failing CI, test, and build logs to surface the single error line, its location, a one-sentence explanation, and a suggested fix. The to…
A developer built ZeroVisual, an open-source visual regression and UI design QA pre-processor that sits between Playwright browser tests and Vision LLMs, cutting token usage by up to 88% and network p…