Sàn agent

Centralized agent control plane for governing runtime agent behavior at scale. Configurable, extensible, and production-ready.

CodeAct Agent SDK with an AST-based code execution engine, enabling stop/start workflows, tool-level state management, and enhanced observability.

Source-available, self-hosted platform for building AI agent flows - visual editor over a Django backend.

She's the AI agent you come home to.

功能那么强大，代码如此简单

Proper sandboxing for agentic coding and web browsing

Continuous CBS - a modification of conflict based search algorithm, that allows to perform actions (move, wait) of arbitrary duration. Timeline is not discretized, i.e. is continuous.

Towards Large Multimodal Models as Visual Foundation Agents

Claude Code 的开源替代方案。Multi Agent AI 编程助手，CLI & Web UI 双入口。自托管，随处部署——手机浏览器也能完整操控。支持任意 LLM，原生浏览器自动化，90+ 命令。技能、插件与持久记忆。企业级安全，零锁定。Docker 一键部署。

Official repository for Agent Harness for Large Language Model Agents: A Survey — 168 references, 34 evidence-coded systems, and 30 harness benchmarks.

C2S Agent is an lightweight AI Agent construction platform that provides configurable online Agents and MCP services, You can configure any HTTP request interface as an MCP tool. C2S Agent 是一个轻量级的AI Agent构建平台，提供在线可配置的Agent，MCP，您可以一个HTTP请求的接口配置成为一个MCP工具，Agent之间可以进行自交流。并提供了单端口多A2A服务，MCP服务的解决方案

Open-source sandboxes where coding agents build and deploy. Spin up isolated environments where Claude Code, Cursor, and other agents code and deploy software.

Remote approvals, policy checks, and execution evidence for unattended AI agents.

Phi-3.5 for Mac: Locally-run Vision and Language Models for Apple Silicon

Agentic Development skills behind the JS Mastery workflow

Mimir - Fully open and customizable memory bank with semantic vector search capabilities for locally indexed files (Code Intelligence) and stored memories that are shared across sessions and chat contexts allowing worker agent to learn from errors in past runs. Includes Drag and Drop multi-agent orchestration

Multi-Faceted AI Agent and Workflow Autotuning. Automatically optimizes LangChain, LangGraph, DSPy programs for better quality, lower execution latency, and lower execution cost. Also has a simple agent/workflow framework

Autonomous multi-branch research lab. Branches compete for compute budget. The system converges on what works.

Python functions powered by AI agents - with runtime post-conditions for reliable agentic workflows.

🐔 Agentic systems explained with chickens — the 2024 taxonomy + the 2026 canon (context engineering, skills, MCP, harnesses). Every pattern as a runnable, CI-checked file. Mermaid everywhere.

🔥[NeurIPS'25] DeepFund: Pilot for Your Next Fund Investment

SkillsVote: Lifecycle Governance of Agent Skills from Collection, Recommendation to Evolution

An introduction to the world of AI Agents

[NeurIPS 2024] ReEvo: Large Language Models as Hyper-Heuristics with Reflective Evolution