Search

Reversible distil context compression for LangChain and LangGraph

A plugin framework for agents, built on LangGraph deepagents + langgraph-cli.

Self-hosted, fail-closed approval server for AI agents (Arbiter)

A topological and physical framework for continuous multi-agent systems.

Testing framework for headless multi-agent LLM pipelines — capture, snapshot, mock, verify.

LangChain Deep Agents adapter for NeMo Fabric

Field kit for engineers embedded in client work - a real CLI (recon, memory, portfolio), one @fde skill with field judgment on top, and hooks that make it automatic. Claude Code plugin and any agent that loads skills.

Agent Skill that teaches AI coding agents (Claude Code, Cursor, Copilot, Codex) to build, debug, and publish browser extensions for Chrome, Edge, and Firefox: Manifest V3 gotchas, cross-browser manifest rules, and live-browser verification playbooks. Powe

ZYR Agent — AI coding agent powered by Google Gemini & OpenRouter

CWS agent runtime SDK — cws-comm protocol layer (WS/auth/heartbeat/reconnect, sync, message codec, tm/kb/as/comm/core/conn service clients) extracted from zylos-openmax; consumed by runtime adapters.

Read-only static scanner for AI-agent codebases: finds dangerous action sinks, proves reachability, attributes guard coverage, and rates findings honestly (reachable-verified vs install-liability).

Peer AI agents coordinate through markdown messages in shared channel folders — no orchestrator, autonomous zero-token waiting, cross-platform.

A Python toolkit for building citation-faithful Q&A agents over a knowledge corpus.

A bilingual, continuously updated directory of AI agent cases, news, open-source projects, papers, and reusable resources.

A free virtual midi keyboard controller for agentic IDEs. Inspired by the Codex Micro and Work Louder physical device. Usable with Codex, Claude Code, BuilderStudio and many others.

Ultra Agentic is an open catalog of Model Context Protocol (MCP) servers, reusable agent skills, and structured datasets for composable AI workflows. This repository contains downloadable beta artifacts, remaining planned roadmap entries, and the static website that presents the catalog.

Frontier models sell confidence. FABULA ships proof — an agent harness where any model is a swappable chip and every finished run mints a replayable, context-fingerprinted Proof-of-Done receipt. Sovereign, local, air-gapped-capable. macOS · Apple Silicon (today).

基于数据驱动与工具调用的多智能体电商运营平台，ERP系统插件，辅助服装店商进行运营决策。基于自主构建的小红书穿搭知识图谱，满足选品需求。

Deterministic personality fluctuation for AI agents and game NPCs — equations, not RNG, with optional BSV proof

Cross-Agent Token Deduplication Bus — eliminate 65-82% of redundant tokens across multi-agent LLM pipelines