A developer benchmarked six models across three labs to test whether LLMs independently verify the arithmetic in startup pitch claims, finding that all three frontier models scored 3/3 with zero varia…
A developer released wconnect, a Python library whose auto_save_in option automatically writes incoming Telegram media to a local directory before a handler function is invoked, removing the need for …
A quality engineer describes using four Model Context Protocol integrations — Jira, Figma, GitLab, and Playwright — to let an AI assistant assemble UI regression tests from manual test cases, design s…
A developer has merged the CHARFORGE character generator and 3D game engine into a single system that supports four distinct paths for getting actors into a game — morph-based human sculpting, importe…
WorldScript Studio, an open-source writing studio, is architected so its AI features are optional rather than load-bearing: all AI capabilities route through a single provider seam (Gemini, OpenAI, Op…
A developer has released Godmode Bot, an open-source (MIT) AI coworker that pairs a real browser with stored logins and 2FA codes while keeping those secrets hidden from the underlying model. Built wi…
A developer released Laya, an Apache-2.0 non-autoregressive decision engine that replaces LLM-as-a-judge prompting with a single forward pass through an encoder, answering typed questions (choice, sco…
A developer building a travel site implemented a defense against invisible prompt-injection attacks by sanitizing all user text before it reaches an LLM, stripping Unicode Tag block characters, zero-w…
A developer built Claude Usage Companion, a free, open-source browser extension that predicts when a user will hit Claude's 5-hour session and 7-day weekly usage limits. The extension captures undocum…
A lead QA engineer used Claude Code as an autonomous operations partner to track down a native memory leak in a JNI bridge between a C server and a Java module, after static code inspection and Valgri…
A developer built two real VS Code extensions using vsceasy, a CLI and framework that scaffolds projects with a React webview, a typed RPC bridge between the extension host and webview, and file-based…
A developer argues that TypeScript's `as` type assertion silently suppresses compiler errors and can ship runtime bugs, and recommends the `satisfies` operator introduced in TypeScript 4.9 instead. Th…
A developer building client chatbots, image generators and voice-cloning tools argues that most AI inference runs on US-based providers such as Replicate, fal.ai, OpenAI and Together, exposing Europea…
An engineer benchmarked roughly 4,900 prompt sessions across five Claude models and seven local Ollama models to measure how spelling, grammar and punctuation errors affect LLM accuracy. Heavy misspel…
A developer demonstrated a stateful PHP agent that reads and mutates inventory, using NanoAgent's FunctionTool to wrap a ProductDatabase class with search_products and place_order tools. The agent que…
A developer demonstrated a retrieval-augmented generation (RAG) pattern in PHP using NanoAgent, in which a knowledge-base search is exposed to an LLM as a callable tool rather than relying on the mode…
A developer built Cubicle, an open-source dashboard that renders AI agents as characters in a pixel-art office, using a single Node server under 200 lines with no dependencies and a 352×208 canvas dra…
A developer built LemonLink, a tool that generates shareable links for any URL and reports whether a human or an AI agent opened them and when. The project targets link analytics for generic links, wh…
An independent research lab, Transluce, published six months of public traces showing AI agents attributed to OpenAI probing a university library, a statistics portal and a public health website throu…
A developer built Sahayak, a RAG-based legal assistant that answers tenancy, consumer-rights and contract questions and summarizes uploaded PDFs in plain language, grounding every response in retrieve…