Tech News
What Is a Harness?
A Hacker News discussion on the concept of 'harness' for LLM agents, where practitioners share insights on building internal CLIs and skills for controlled agent workflows, highlighting the gap between terminology.
llm-openrouter 0.7
Now that this plugin is compatible with LLM 0.32 it can display the reasoning traces for LLMs available through OpenRouter. Updated for compatibility with LLM.
Agentic Data Operations Platform (ADOP): Data engineering into hours
The Agentic Data Operations Platform (ADOP) is a reference architecture on Amazon Bedrock that uses specialized AI agents to automate the full Bronze-to-Silver-to-Gold data pipeline lifecycle, compressing new-source...
NVIDIA AVO Reaches 100% on ARC-AGI-3, Demonstrating a Frontier-Level General-Purpose Architecture for Long-Horizon Autonomous Agents
A frontier language model is only one component of an AI agent. The surrounding agent system—often called a harness—determines how the model receives.
I spent $266 and four AI models to own my tablet. GLM-5.3 finished it in a day
A developer spent $266 and used four AI models to fully own and customize a Fire HD tablet, with GLM-5.3 completing the task in a day.
GitHub Repos
Panniantong/Agent-Reach
Give your AI agent eyes to see the entire internet.
# Python# agent infrastructure# ai agentKunAgent/Kun
Local-first AI agent workspace for coding, writing, design, research, and automation — one runtime for desktop GUI and TUI.
# TypeScript# agentic workflow# ai agentbrowserwing/browserwing
BrowserWing turns your browser actions into MCP commands Or Claude Skill, allowing AI agents to control browsers efficiently and reliably.
# Go# aibrowser# golangResearch Papers
Graph Engineering in the Era of LLM Agents: From Individual Intelligence to System Intelligence
This paper introduces 'Graph Engineering' as a new paradigm for designing LLM agent systems, where tasks, agents, and states are organized as dynamic graphs to achieve 'System Intelligence'—coordinating multiple specialized agents to solve complex problems that exceed...
EnSI-RAG: Entity-Structure-Indexed Retrieval-Augmented Generation for Long-Document Question Answering
This paper introduces EnSI-RAG, a retrieval-augmented generation method that builds an entity-centered index to improve question answering over long documents. It stores entity properties, relations, and aspects as structured records, which helps retrieve relevant evidence...
SRL-MPC: Shape-Aware Reinforcement Learned Model Predictive Control
This paper introduces SRL-MPC, a method that helps robots navigate through crowded spaces while considering their own shape and the shapes of others, without simplifying geometry. It uses reinforcement learning to adjust control parameters in real time, improving safety and...