Persistent memory system for AI agents - remember everything across sessions with a simple 426-token prompt
Build, run, and manage AI agent platforms with full control over your data and infrastructure
Python SDK for connecting LLMs to external tools and data via Model Context Protocol
Infrastructure-as-Effects — cloud infrastructure and application logic as a single, type-safe Effect program
Google's declarative runtime for orchestrating billions of autonomous AI agents at cluster scale
Connect AI agents to Cloudflare services using natural language through the Model Context Protocol
Durable execution framework with built-in AI coding assistant support for resilient applications
Connect private MCP servers to ChatGPT and Codex without exposing them to the public internet
Grafana CLI optimized for AI agents — query production, investigate alerts, and ship fixes without leaving your editor
Open-source runtime powering AI agent cloud with secure Firecracker microVMs for code execution
Distributed AI compute mesh - pool GPUs across machines for shared LLM inference
Lightning-fast trigram-indexed search engine powering GitHub Copilot CLI for instant codebase searches
Local-first long-term memory runtime for AI agents with markdown storage and SQLite indexing
Stateful API layer for agentic applications - run OpenAI-grade workloads on your own GPUs with vLLM
Open source gateway and router for AI agent workflows with unified model access and spending controls
Fast, isolated microVMs for AI agents with <100ms boot time — run agents on real machines with their own kernel and filesystem
AI agents for your UniFi infrastructure - manage networks, cameras, and access control with agentic workflows
Self-hosted deployment platform with AI-powered auto-configuration and zero-config shipping
High-performance Kubernetes platform for scaling AI agents and autonomous workloads
Remote testing and execution control plane for AI agents and development teams
Model Context Protocol server that gives LLMs native Kubernetes and OpenShift cluster management capabilities
Lightning-fast, secure sandbox runtime for AI agents and code execution
Enable GPU acceleration in Kubernetes clusters for AI/ML workloads
Making all software agent-native — bridge AI agents to any application with auto-generated CLIs
Equip AI coding agents with AWS deployment and architecture skills through reusable plugins
Safe, private runtime for autonomous AI agents with sandboxed execution and declarative security policies
High-performance AI API proxy gateway for Claude, OpenAI, Codex, and Gemini with unified management
Self-hosted cloud development environments and AI coding agents with centralized governance
AI-powered open source real-time observability and monitoring system
AI-agent-ready CLI with 200+ commands for Datadog's observability platform
The open-source notification infrastructure for multi-channel delivery in a single component.
Kubernetes infrastructure for managing isolated AI agent runtimes with stable identity and persistent storage
Self-hosting platform for Docker apps with AI agent CLI support
Modern web interface for Nginx configuration and management
Single-file memory layer for AI agents with instant retrieval and long-term memory
自建Claude Code镜像服务，支持拼车共享的中转API工具
