Baldur, an early-access Python framework from baldurhq, released a decorator that makes LLM workers share a single backoff wait on a 429 or "overloaded" response and parks failed jobs for replay inste…
Developer NathanAB released statusline-anywhere, a free MIT-licensed Claude Code plugin that renders custom status lines inside the Claude Desktop app, which does not natively draw them (issue #41456)…
A developer released "What's Agent Doing," an MIT-licensed Claude Code UI mod that displays the current step, a timer, and per-background-agent status in a single line above the prompt. The plugin use…
Google released Mantis, an open-source kit of skills and an ADK reference harness that lets coding agents autonomously find, triage, reproduce, and patch security vulnerabilities in codebases. Mantis …
Arrowproof, an agent skill released by developer ahmtsahin, verifies every box and arrow in an LLM-drawn Excalidraw architecture diagram against a repository's real import graph. In a Flask test, Clau…
A developer released ScopeLock, an open-source Git pre-commit hook and CLI tool that scans staged diffs for scope-creep keywords, converts out-of-scope developer hours into dollar variance at a defaul…
A developer published a handover guide for running local LLM inference on an AMD Strix Halo machine (Ryzen AI Max+ 395, gfx1151, 128 GB unified memory), configuring LlamaStash to launch Qwen3.8 Flash-…
A new open-source tool called jev maps the attack surface of Python, JavaScript and TypeScript codebases and flags likely vulnerabilities down to the exact line for about one cent per repository, usin…
Hopsworks is hosting two self-hosted decision models, qwen and kumo, that play the procedural game Hops Run live at game.hopsworks.ai, with each move sent to a Hopsworks model deployment that returns …
Developer amergrgic released Kodama, an independent community multi-agent pack for the Kiro IDE and Kiro CLI, installing 9 agents and 3 skills via a clone-and-install script or a curl one-liner. The p…
Free Coding Agent has launched as an open-source, provider-neutral MCP execution layer that gives any MCP-capable frontier model controlled access to files, terminals, Git, language servers, browsers,…
Developer rav4nn released agent-chrome, a free open-source Claude Code plugin that lets the AI agent browse in a copy of the user's signed-in Chrome while the browser window stays behind the terminal …
ImageGen, a native macOS app from developer ph1lb4, runs Alibaba's Qwen-Image 2.1 image model fully on-device on Apple Silicon, requiring about 33 GB of unified memory and roughly 35 GB of free disk, …
AgentTerm, an MIT-licensed terminal that expands each AI CLI session into its own visual workspace, has been released for macOS and Windows via WSL with no telemetry or account required. The developer…
Diffusion Studio released a Codex plugin, published as a private GitHub repository at Teseife/diffusion-studio-codex, that lets users create and edit video projects from OpenAI's Codex client and cont…
A developer published funcroute, a roughly 4,000-line C router that sits between a client and model vendors and routes each request to a specialist upstream based on its attachments, letting a cheap t…
Developer lox released agent-harness, a minimal, composable Go library (requiring Go 1.26+) for building agentic tool-calling loops on top of LLM APIs. The library implements the core agent loop — cal…
Insta Fusion released Better Call GPT, a free MIT-licensed Claude Code plugin that lets developers talk to their coding session through a realtime voice model, with the voice layer unable to approve p…
Braxis launched as a Python tool that generates and keeps four AI agent context files — AGENTS.md, CLAUDE.md, .cursorrules, and .agentic-config.json — in sync with a codebase via the `braxis generate`…
A developer released Visi, an open-source command-line spreadsheet engine built with LLMs and fuzzing that aims for near-100% Excel compatibility, according to a Show HN post. After a couple of months…