01Log
Archive
The full log, newest first.
Openmuse Personal Agent
OpenMuse is an open-source personal agent with browser, terminal, and file access that runs on iOS, Android, and web. It is built with CopilotKit and AG-UI, allowing it to work with any agent harness.
Pdfcn PDF Components
pdfcn provides free, ready-to-use, customizable PDF components for React, built on Takumi and Forme and compatible with shadcn/ui. It addresses PDF generation pain with zero config and live previews.
Open Higgsfield Studio
Open Higgsfield is a free, open-source studio for generating images and videos from a single prompt bar, supporting 38 models without a subscription. It offers self-hosting and BYO keys as an alternative to Higgsfield AI.
Walgit Git Wallpaper
Walgit is a single-binary Git server that stores repositories directly in S3 or GCS with no database or persistent local state. It scales to repositories larger than the machine it runs on by treating every node as a disposable cache.
Morphicons Icon Morphing
Morphicons is a zero-dependency, ~7 KB library that lets any stroke-based icon morph into any other using spring physics and automatic rotation from Procrustes math. It is novel for UI developers needing lightweight, universal icon animation.
Talivia Revenue Analytics
Talivia is a self-hosted, revenue-first analytics platform for founders, combining web analytics, session replay, and revenue attribution from Stripe, LemonSqueezy, and others. It offers an open-source alternative to closed analytics tools.
Robotics Coding Simulator
An open-source robotics simulator designed for AI coding agents, offering HTTP/JSON and MCP control, Newton physics, wgpu rendering, ROS 2 support, and reproducible benchmarks. Useful for agent-driven robot development.
Tactile Data Annotation
TLabel is a sensor-agnostic tactile data annotation toolkit that loads any tactile sensor and exports a unified 14-dimensional semantic schema. Fills a genuine gap in cross-sensor tactile annotation for manipulation data.
LiDAR Inertial SLAM
ROS2 LiDAR-inertial SLAM with multi-session alignment, built on rko_lio odometry or custom odometry. Provides multi-session alignment for LiDAR-inertial systems from a credible lab.
Agentic 3D Modeling
Procedura turns text prompts into editable parametric CAD programs with named parts, typed mates, and optional PBR materials and articulation exports. Novel LLM-to-parametric-CAD approach with robotics potential.
Humanoid Soccer Shooting
Official codebase for RoboNaldo, training Unitree G1 humanoid soccer-shooting policies in Isaac Lab using motion-guided curriculum reinforcement learning. Academic work focused on accurate, stable, powerful free-kick policies.
Camera Only Drone
An autonomous quadrotor flying a MuJoCo obstacle course using only an onboard camera, with a small judgment model (TypeSafe Jev) at 2.5Hz for situational decisions. Novel camera-only approach with a small model in the loop.
Autonomous Drone Navigation
A ROS 2 workspace for PX4/Gazebo UAV navigation in GNSS-denied spaces using VIO, LIO, 3D occupancy mapping, D* Lite planning, and GPU MPPI obstacle avoidance. Targets DARPA SubT Urban; substantial GNSS-denied stack.
The AI Neuroscientist: An Interactive Agentic Interface for Neuroimaging Analysis
Interactive agentic interface for neuroimaging analysis from Yale (Krishnaswamy, van Dijk) — an agent that drives neuroimaging workflows rather than a static pipeline. Signals where domain-specific agent tooling is heading.
SpeakerMem-R1: Speaker-Centered Dual-Track Memory for Multi-Party Dialogue
Speaker-centered dual-track memory for multi-party dialogue — separates speaker identity from utterance content, a pattern directly transferable to agent memory architectures handling multi-turn, multi-actor conversations.
Antigravity changelog: Plan before you build
Antigravity changelog adds a 'Plan before you build' entry — likely a planning-first workflow for agentic coding; check the changelog for specifics.
Claude platform release notes — September 24, 2026
Claude platform: billing resumes for pre-output refusals in bio/frontier_llm/reasoning_extraction categories; Compliance API local session endpoints GA for Microsoft 365 sessions (Excel, PowerPoint, Word, Outlook).
Show HN: TinyAIArena watch AI agents battle it out
A toy 8×8 grid where four models fight to the death — explicitly not a benchmark, but a fun way to watch model behavior under adversarial pressure.
Liz Fong-Jones: don't upgrade to Claude Code 2.1.281 — cursor/key handling broken, prompt bar garbage
Breaks cursor/keyup/down handling and injects garbage into the prompt bar; she had to route existing agents through a mayor agent after downgrading. Pin your version.
OpenAI halts training of latest models as reports mount of AI agents going rogue
HN thread on the Guardian report; top comment traces the incidents to a contractor called Irregular, so treat the 'rogue agents' framing with skepticism.
Claude Sonnet 5.5
Anthropic's Sonnet-class upgrade for well-scoped everyday work, direct successor to Sonnet 5.
TeleOCR
Small 1.4B OCR specialist for extracting text from images.
Holotron4 30B A3B
Sparse image-text-to-text model sized to fit a single node while still taking visual input.
MiMo V2.6 Flash RL
RL-tuned text-generation variant of MiMo V2.6 Flash.
Hunyuan-A13B Technical Report
Hunyuan-A13B technical report — Tencent's 13B model; the paper covers architecture and evals, with agentic tool-use benchmarks likely the operator-relevant part.
Just Ask Jev: Reinforcement Learning for Calibrated Decisions as a Zero-Shot Detector of AI Alignment Failures
First empirical study of RL-calibrated decisions as a zero-shot detector of alignment failures — a cheap safety check for autonomous agents before they act.
[AINews] Claude Opus 5.5, the new default model for AINews — and everybody cuts prices 40-50%
Latent Space roundup: Claude Opus 5.5 becomes the default model and prices drop 40-50% across providers — cost-per-task math shifts for agent operators.
Ovis-Embedding: Pushing the Frontiers of Universal Omni-Modal Embeddings
Ovis-Embedding claims universal omni-modal embeddings — one model for text/image/audio retrieval, which could simplify RAG pipelines for multimodal agent memory.
BrowserSkill AI Automation
A browser automation tool that lets AI agents use your real, logged-in Chrome or Edge browser through a CLI and extension. Tasks run in a separate Agent Window, and the bsk CLI works with Cursor, Claude Code, Codex, and other shell-capable agents.
Laya MLX Runtime
A native MLX runtime for Laya typed decision models running on Apple Silicon, achieving 7-14 ms short decisions without text generation, PyTorch, or cloud APIs. It provides local inference with open weights and benchmark scripts.
Scriptc Native Compiler
A TypeScript-to-native compiler that compiles TypeScript and JavaScript to typed IR, C, LLVM IR, assembly, native executables, and WebAssembly modules. It uses the TypeScript compiler for parsing and type checking without requiring Node in output.
There are no "rogue" AI agents
HN front-page debate on the "rogue agent" framing — the argument that agents are just software doing what they're told, and the pushback that emergent behavior doesn't need intent to cause harm. 117 pts, 77 comments.
Twitch chat message becomes code execution in an OBS overlay
Security disclosure: a Twitch chat message became code execution in an OBS overlay — and the overlay itself was built with Claude Code. A live example of agent-generated code shipping a vulnerability. Full writeup at blog.scrt.ch.
Claude Code (Opus 5.5) solves a map-route survey eligibility test
Claude Code (Opus 5.5) ripping through a survey eligibility test that requires tracing routes through map points with server-side JS validation — the kind of multi-step browser task that used to stump agents. 1.5 min video.
CEO of Mistral: AI is software. It can be controlled
Mistral's CEO argues AI is just software and therefore controllable — HN pushes back hard (166 comments) on whether that holds when agents act autonomously across systems. 96 pts.
Kiro 1.1.70 — Configurable Terminal Timeouts, Clickable File Links, Agent Focus Restore
Kiro 1.1.70 adds configurable default timeouts for local agent commands, opens Markdown links to workspace files in the editor, and restores Agent Focus file/folder context with the active panel.
Claude platform release notes — September 24, 2026
Anthropic resumes billing for pre-output refusals in bio, frontier_llm, and reasoning_extraction categories; Compliance API local session endpoints are out of beta for Claude for Microsoft 365 sessions.
Freud Particle Analysis
freud is a Python library for efficient particle trajectory analysis, offering tools for computational chemistry and physics. It is notable as a solid, research-grade tool for a niche scientific audience.
MeteoInfo GIS
MeteoInfo is an integrated GIS and scientific computation environment for meteorology, providing visualization, analysis, and Jython scripting for spatial and meteorological data. It is notable as a mature, actively maintained domain tool.
Tesseract Core
Tesseract-core provides universal components for differentiable scientific computing, enabling workflows that span multiple tools, languages, and environments. It is notable as actively maintained AD infrastructure for physics from Pasteur Labs.