The Center for AI Standards and Innovation (CAISI) released a repository packaging 2 agents and 3 benchmarks for agentic cyber capability evaluations of AI systems, built on the Inspect evaluation fra…
RehanMohammed985 released Runtape, an open-source Python tool that performs counterfactual debugging and regression testing on AI agent runs, installable via pip for Python 3.10+ and compatible with t…
A from-scratch Rust language model called PSSA, which replaces transformer attention with a recurrent state-space core and a 512-slot episodic memory bank, reached 3.98 training cross-entropy versus 4…
Moching, an open-source desktop AI agent built in Rust by moching-ai-dev, launched on GitHub with 219 built-in tools spanning PC system control, browser automation, office documents, media processing,…
A developer documented a workaround for rendering SwiftUI previews on chosen devices and orientations by driving Xcode's MCP bridge (xcrun mcpbridge) from an external agent on Xcode 27.0 RC. Orientati…
An open-source rule set called data-agent-rules stopped destructive AI agent behavior in a 27-run eval, after Claude Haiku 4.5 with no rules wiped every row of prod.orders while "removing duplicates,"…
A developer released Image Categorizer, an open-source tool that sorts a photo library into categories using local vision and text models served by Ollama, keeping all photos on the user's own network…
Microsoft released WSL 3.0.1, marking the general availability of WSL containers (WSLc) for running Linux containers on Windows, including for AI workloads. The release notes list 10 changes, among th…
Developer Ian K. released fab, an open-source browser CLI for AI agents, claiming 92.5% task pass rate versus 73.1% for LLM plus chrome-devtools-mcp, at a median 16.6 seconds per task versus 40.7 seco…
OpenAI released the Sign in with ChatGPT DevKit, a local Node.js SDK and React component set that lets developers add ChatGPT account sign-in to open-source apps running on a user's own machine and, w…
A developer released TurboGPT, an MIT-licensed byte-level GPT trainer written in CUDA C++ that trains a 22KiB transformer in 13 seconds. The tool builds on Windows with Visual Studio 2022 C++ tools an…
OpenJev, an independent open-source project from GPT-AGI, released a Jev-compatible System One decision engine that answers typed questions (Choice, Score, Noul) from open-weight models in a single fo…
Developer Simar-malhotra09 released pi-sheep v1.0.0, a pi extension installable via `pi install git:github.com/Simar-malhotra09/pi-sheep@v1.0.0` that renders a sheep animation when an agent runs a sle…
An independent developer released Open Dots, an open-source, self-hosted AI workspace positioned as an alternative to OpenAI Dots, offering chat, a deny-by-default action gateway with approval prompts…
DiggerHQ released OpenDots, an open-source, model-agnostic personal agent framework positioned as an alternative to OpenAI's Dots, running on OpenComputer Serverless Agents. OpenDots is a TanStack Sta…
Kompyla is a self-hosted research monitor that treats an LLM like a compiler, turning raw documents into structured, cited wiki pages. The tool extends Andrej Karpathy's LLM Knowledge Base pattern wit…
OpenZL released v0.3.0, whose new LZ engine uses Marcin Żukowski's PivCo Huffman entropy backend to cut decompression time by 33% over v0.2.0 at compression level 1 with a 64KB window, reaching 3062 M…
Estela, a new open-source CLI from developer jdleonruiz, rebuilds billable hours and AI costs by reading the session logs Claude Code writes to ~/.claude/projects/, Codex writes to ~/.codex/sessions/,…
Phaze released a working example on Hacker News's Show HN of an IT helpdesk agent built on the Phaze Enterprise API and Phaze MCP server that watches a Slack channel, locates a user's Windows machine,…
A developer known as Ahmed produced an eight-minute animated film using Claude Sonnet 5.5 inside Claude Code, orchestrating one sub-agent per scene through a shared HyperFrames HTML-to-video pipeline.…