Search

Production-grade multi-agent harness with LLM routing, memory, safety, and observability

AI model: internlm/Agent-FLAN-7b

AI model: scrubster/dr-stein-colab-qwen15-tooluse-sanity

Production-grade multi-agent harness with LLM routing, memory, safety, and observability

Local-first AI agent framework for personal workflows and enterprise support, built in Go with memory, knowledge base, MCP, plugins, webchat, WeChat/WeCom channels, and self-update.

AI Agent 驱动的 Primavera P6 施工进度计划生成工具 | CSV → XER → P6

An AI agent framework built on a cat's biological blueprint. Define your organs, wire their nerves, and the cat comes alive.

Toma control de tu Google Ads, GTM y GA4 desde la terminal — kit open source con playbook para Claude Code.

PHP 8.2 MVC starter template with PHPStan, Pest, Gulp/Vite, AI agent support and full tooling

Final Project for the course - UIUC CS 498: AI Agents in the Wild (Spring 2026). This repo contains RescueBench, a 20-scenario benchmark for emergency-response resource allocation, and a modular agent framework for deterministic and LLM-assisted dispatch, simulation, and evaluation.

Hermes Agent on a Dedalus microVM, with a bundled skill library, scheduled crons, a Reticle-styled chat UI, and the Cursor TypeScript SDK wired in as an MCP tool so the agent can spawn coding agents that inherit the rig's conventions as .cursor/rules.

DeepSeek-native AI coding agent for your terminal. Engineered around prefix-cache stability — leave it running.

Run your own agents. Pluggable claude-codes, and multi-channel, your transcripts stay local. 🪄

Your agent's librarian - the sidekick MCP server for context management to any agent

Assignment 3: Coding LLM Agents – EN.601.727 Machine Programming

Architectural decision records on accountability distribution in autonomous AI agents — 10 ADRs + 4 Business AI Quadrants + Phase Separation axis. Harness-neutral practice library extracted from contemplative-agent. DOI on Zenodo.

AI context observability. See what your LLM saw.

CLI-Toolkit für AI-Agent-Infrastruktur: .agent-Ordner, Tool-Discovery, Credential-Management

ISOPro: a reference implementation of Grounded Continuous Evaluation (GCE) — simulation-based fine-tuning and evaluation for LLMs. Replaces the learned reward model with a deterministic verifier and updates LoRA adapters on CPU, eliminating reward hacking by construction in verifiable-reward domains.

Install dot-skill (colleague.skill) into your AI agent host. Distill anyone into a reusable AI Skill.