Awesome GitHub RepositoriesRequest Filter Integrations

This project provides a system for managing agent context and session memory, featuring an agent context compactor, an AI session memory manager, and a tool output sandbox. It functions as a middleware layer and server extension for the Model Context Protocol to optimize context windows and reduce token usage. The system optimizes agent performance by sandboxing tool outputs and externalizing large data sets, replacing raw I/O with pointers and concise summaries. It employs a persistent knowledge base that indexes session history and tool outputs for retrieval via full-text search, ensuring s

Presto is a distributed SQL query engine designed for high-performance analytical processing across heterogeneous data sources. It functions as a data federation platform and massively parallel processing engine, allowing users to execute interactive queries against diverse storage systems without requiring data migration. By mapping remote metadata and structures to a unified relational namespace, it enables seamless cross-platform analysis through a standard SQL interface. The engine distinguishes itself through a pluggable connector architecture and a shared-nothing distributed processing

Quarkus is a Kubernetes-native Java framework designed for building high-performance, memory-efficient applications. It utilizes ahead-of-time native compilation to transform Java code into standalone, optimized binaries that eliminate the need for a virtual machine, enabling rapid startup and reduced memory consumption. By performing code augmentation during the build phase, it shifts heavy processing tasks away from runtime, ensuring that applications are optimized for cloud-native environments. The framework distinguishes itself through a unified approach to reactive and imperative program

Tengine is an enhanced Nginx web server distribution designed for high-traffic environments. It functions as a high-performance reverse proxy, dynamic load balancer, and HTTP/3 proxy server, integrating an embedded Lua scripting engine to allow for custom business logic and dynamic request processing. The project distinguishes itself through a focus on modern transport protocols and efficiency, featuring native support for QUIC and HTTP/3 to reduce connection latency. It implements userspace UDP forwarding to bypass the kernel for higher throughput and provides the ability to dynamically upda

Orleans is a .NET distributed actor framework designed for building scalable, cloud-native applications. It implements a virtual actor model where entities with stable identities manage their own state and lifecycle across a cluster of servers. The framework provides a distributed state management system with ACID transaction support and a distributed pub/sub streaming engine for real-time data processing. It distinguishes itself through location-transparent routing, automatic actor activation and deactivation, and elastic cluster scaling that redistributes workloads during node failures. Th

Healthchecks is a heartbeat monitoring service and cron job monitoring tool designed to track the execution and success of scheduled tasks and systemd timers. It functions as a dead man switch, alerting users when expected periodic signals from remote processes fail to arrive. The system accepts health signals via HTTP and SMTP, allowing it to track infrastructure heartbeats from sources ranging from CI/CD workflows to network routers. It distinguishes itself by supporting the capture of diagnostic data, including exit codes and execution logs, and by calculating the duration between start an

Plano is an AI agent orchestrator and LLM gateway proxy that unifies access to multiple AI providers through a single interoperable interface. It functions as a model routing engine that decouples applications from specific vendors using semantic aliases, allowing traffic to be shifted between providers without modifying application code. The system distinguishes itself with intent-based agent routing, which directs prompts to specialized agents based on semantic analysis. It features an interceptor-based filter chain system that acts as guardrail middleware to enforce safety policies, rewrit

MagicOnion est un framework RPC .NET utilisé pour construire des services d'appel de procédure à distance typés en utilisant des interfaces partagées pour la communication entre serveurs et clients. Il fonctionne comme un framework de sérialisation binaire et un orchestrateur de service distribué, fournissant une couche réseau pour le streaming à faible latence et l'échange de données en temps réel. Le projet se distingue par un client réseau compatible AOT qui utilise des générateurs de code source pour prendre en charge la compilation ahead-of-time dans des environnements d'exécution restreints. Il permet une communication bidirectionnelle et le streaming de données en temps réel, permettant aux serveurs et à plusieurs clients de pousser des messages de manière asynchrone sans polling. Le framework couvre une large surface de capacités, incluant la gestion de connexion avec état basée sur des sessions, des pipelines de requête basés sur des intercepteurs pour la validation, et le contrôle d'accès basé sur les rôles. Il inclut également des outils d'observabilité tels que la surveillance de connexion basée sur le heartbeat et les métriques de performance du serveur, aux côtés de la prise en charge de la communication réseau Unity.

MagicOnion est un framework d'appel de procédure à distance pour la plateforme .NET et Unity. Il fonctionne comme un protocole d'API binaire et un moteur de communication en temps réel qui utilise des interfaces partagées pour fournir une communication typée entre clients et serveurs. Le framework se distingue par sa compatibilité avec les environnements de compilation ahead-of-time, utilisant des générateurs de code source pour les proxies clients et la sérialisation pour prendre en charge les plateformes mobiles et consoles, y compris celles utilisant IL2CPP. Il fournit une couche middleware spécialisée pour Unity qui gère les types de données spécifiques au moteur et assure la compatibilité à l'exécution. Le projet couvre un large éventail de capacités réseau, incluant le streaming de données bidirectionnel, les cycles requête-réponse unaires, et la messagerie push serveur-vers-client. Il inclut une prise en charge intégrée de la sécurité via le chiffrement TLS et l'authentification JWT, ainsi qu'une gestion de session avec état par client et un filtrage de requête basé sur des intercepteurs. Des outils de diagnostic sont disponibles pour surveiller l'état de la communication, suivre les métriques de performance du serveur, et propager les traces de pile côté serveur aux clients pour le débogage.