Awesome GitHub RepositoriesSession State Persistence

OpenHarness is a framework for building and orchestrating AI agents that utilize tools and plugins to execute complex tasks. It provides an orchestration system for managing language model lifecycles and a multi-agent coordination system for delegating workloads across teams of specialized subagents. The project features an agent gateway that bridges language model agents to external chat platforms and communication channels. It includes a tool integration engine for executing shell, file, and web operations, supported by a memory and skill manager that handles persistent user preferences and

lnav is a terminal-based log viewer and analyzer designed for aggregating, filtering, and analyzing multiple log files in a single chronological view. It functions as a console application that can replace the system pager, providing syntax highlighting and document navigation for system or application logs. The project distinguishes itself by mapping unstructured log data to virtual SQLite tables, enabling the use of SQL and PRQL for structured data analysis, aggregations, and relational queries. It further differentiates its capability set through native integration for retrieving and taili

http-prompt is an interactive HTTP client and API testing CLI designed for constructing and debugging network requests to REST and GraphQL endpoints. It functions as a command-line request builder and a shell extension built on top of HTTPie. The tool provides an interactive shell interface featuring request autocomplete and real-time syntax highlighting for requests and responses. It incorporates stateful session management to maintain request history and session data across multiple commands. The project covers API request automation and REST API debugging, allowing users to send and verif

This project is a Rust-based AI agent framework and tool orchestrator that provides a command-line interface for interacting with large language models. It functions as an AI tool orchestrator that routes client requests to language servers and manages the planning and handoffs between specialized agents to solve complex tasks. The system distinguishes itself as a language porting validator, using deterministic mocks and specifications to verify feature parity between different language implementations of a codebase. It further extends agent capabilities by acting as a Model Context Protocol

Venom is a TypeScript library for building automated bots that interact with WhatsApp. It provides a complete toolkit for programmatically managing chats, groups, and messages by bridging into the WhatsApp Web protocol through WebSocket communication. The library handles authentication via QR code scanning, session persistence to avoid repeated logins, and real-time event listening for message delivery, edits, deletions, and connection state changes. The library offers a command-based routing system that directs incoming messages to handler functions based on pattern matching, alongside an ev

mistral.rs is an inference engine for large language models that runs locally and exposes models behind OpenAI and Anthropic-compatible APIs. It serves as a multi-model serving platform, capable of loading several models in a single server process with per-request routing and on-demand loading and unloading. The engine supports multimodal inference, processing text alongside images, video, audio, and speech inputs, and includes a quantized model deployment runtime that reduces memory use and speeds up inference on consumer hardware. The project distinguishes itself through an agentic tool exe

TanStack Form is a cross-framework form state management library that provides typed fields, validation, and submission across React, Vue, Angular, Solid, Lit, Svelte, and Preact. It uses a shared form model that adapts to different UI frameworks while preserving the same validation and submission logic, and offers headless form controls that impose no UI markup, letting developers bring their own inputs and design system. The library distinguishes itself through granular state subscription, where components subscribe to narrow slices of form or field state using reactive primitives, so only

This project is a business rules management system and rule engine designed to define, execute, and manage complex business logic decoupled from application source code. It provides a business logic compiler that transforms human-readable rule definitions into executable models for high-performance runtime evaluation. The system includes a complex event processing engine for analyzing real-time data streams to identify temporal patterns and a decision model and notation executor that processes structured logic based on industry standards for deterministic results. It utilizes an inference eng

This is an open-source Python SDK for building and orchestrating production-grade AI agents. It provides a unified framework for creating conversational agents that can use tools, maintain state, and coordinate across multiple language model providers including OpenAI, Anthropic, Google, Amazon Bedrock, and locally-hosted models. The SDK supports multi-agent orchestration through graphs, teams, and swarms, allowing several specialized agents to collaborate on complex tasks. Agents can be composed as callable tools that other agents invoke, and the framework includes policy handlers that inspe

Dieses Projekt ist eine Bibliothek und ein Kommandozeilen-Interface für die Inferenz lokaler Large Language Models. Es ermöglicht die Generierung von Textvervollständigungen und Chat-Antworten verschiedener Modellarchitekturen. Das Projekt bietet Tools für Weight-Quantization, um den Speicherbedarf zu reduzieren, und integriert Hardwarebeschleunigung durch GPU-Offloading, um die Berechnungsgeschwindigkeit zu erhöhen. Zudem enthält es Utilities für die Modellevaluierung durch Messung der Perplexity auf spezifischen Datensätzen. Die Funktionen decken den gesamten Inferenz-Lifecycle ab, einschließlich binärem Modell-Laden, Template-basierter Prompt-Strukturierung und Sitzungspersistenz zur Beibehaltung des Konversationskontexts. Es unterstützt zudem Aufgaben-Orchestrierung, was es ermöglicht, mehrere Modellaufrufe in Pipelines für mehrstufige Operationen zu sequenzieren.

Genkit ist ein LLM-Anwendungs-Framework und ein Toolkit für generative KI, das für die Entwicklung produktionsreifer KI-Anwendungen konzipiert wurde. Es dient als KI-Workflow-Orchestrator, der Modellaufrufe und den Einsatz von Agenten-Tools durch typsichere Ausführungsabläufe koordiniert. Das Projekt bietet eine einheitliche Modellschnittstelle und eine Plugin-Architektur, um den Zugriff auf verschiedene Large Language Models, Vektordatenbanken und Telemetrie-Backends zu standardisieren. Es zeichnet sich durch eine dedizierte Observability-Suite zur Nachverfolgung von Ausführungsschritten sowie ein Entwickler-Toolkit zum Prompting, Debugging und Evaluieren von KI-Logik über eine lokale Schnittstelle aus. Das Framework deckt ein breites Spektrum an Funktionen ab, darunter Agenten-Orchestrierung mit Tool-Calling und Sub-Agenten-Delegation, Retrieval-Augmented Generation (RAG) durch Vektordatenbank-Integration sowie die Generierung strukturierter Ausgaben mittels schema-basierter Validierung. Es enthält zudem Systeme für zustandsbehaftetes Sitzungsmanagement, ereignisbasiertes Response-Streaming und die Möglichkeit, KI-Flows als skalierbare HTTP-Endpunkte bereitzustellen. Die Entwicklung wird durch eine Command-Line-Interface (CLI) zum Ausführen von Funktionen und Verwalten von Logs unterstützt.

Genkit is an open-source framework for building AI-powered applications. It provides a unified interface for connecting to hundreds of generative AI models from multiple providers, enabling text, image, audio, and video generation through a single API. The framework structures multi-step AI interactions—including chat, retrieval-augmented generation, tool use, and agentic workflows—as composable, traceable flows with built-in streaming and state management. The framework distinguishes itself through a comprehensive developer toolkit that includes a command-line interface and a local developer

Reverse engineering and pentesting for Android applications

Commix is an automated tool for detecting and exploiting OS command injection vulnerabilities in web applications. It probes user-supplied input vectors with heuristic test payloads, analyzes response differences to identify injection points, and then automates the execution of arbitrary operating system commands on the target server. The tool distinguishes itself through a multi-layer filter bypass engine that evaluates input constraints independently per filter type and composes tailored evasion strategies into a single payload. A modular payload tamper pipeline transforms raw injection str

Panel ist ein Python-Framework für Datenanwendungen, das zum Aufbau interaktiver Dashboards und reaktiver Benutzeroberflächen durch die Verbindung von Datenvisualisierungen, Widgets und Callbacks verwendet wird. Es fungiert als reaktive UI-Bibliothek, die den Anwendungszustand mit Interface-Updates basierend auf Benutzerinteraktionen synchronisiert. Das Projekt zeichnet sich dadurch aus, dass es mehrere Ausführungsmodelle anbietet, darunter einen FastAPI-basierten Webanwendungsserver für skalierbares Session-Management und einen WebAssembly-Compiler, der es Python-Anwendungen ermöglicht, direkt im Browser ohne Backend-Server zu laufen. Es enthält zudem ein spezialisiertes Toolkit für den Aufbau konversationeller KI-Interfaces mit Streaming-Textantworten und Nachrichtenverläufen. Das Framework deckt breite Funktionsbereiche ab, einschließlich responsivem Layout-Design, der Integration diverser Plotting-Bibliotheken und interaktiver Datentabellen sowie dem Management reaktiver Parameter. Es bietet Werkzeuge für Benutzerauthentifizierung, Session-Persistenz und die Planung von Hintergrundaufgaben.

Vorpal is a Node.js interactive CLI framework and terminal user interface library used to build extensible command-line shells. It functions as an interactive command-line parser that converts string input into executable functions, managing the lifecycle of terminal sessions and command routing. The framework is distinguished by a plugin-based extension architecture that allows external modules to register new commands, shared behaviors, and complete command suites into the core environment. It supports the creation of custom shell environments with specialized namespaces and a system for pe

Dieses Projekt ist eine Bibliothek von Automatisierungsskripten für die BoxJs-Umgebung. Sie fungiert als externer API-Client und Integrations-Interface für die Interaktion mit Remote-Servern. Das System bietet einen persistenten lokalen Datenspeicher, um den Zustand zwischen Skriptausführungen durch das Speichern und Abrufen von Strings und JSON-Objekten beizubehalten. Es ermöglicht den Datenaustausch durch strukturierte JSON-Objekte und unterstützt sowohl synchrone als auch asynchrone Netzwerkanfragen unter Verwendung benutzerdefinierter Header.

Inngest is a durable execution framework and event-driven automation engine designed to orchestrate background workflows. It enables developers to build resilient, stateful processes by memoizing function steps, ensuring that long-running tasks can automatically resume from the last successful operation after failures, timeouts, or infrastructure restarts. The platform distinguishes itself through its event-driven architecture, which uses a schema-validated bus to trigger functions and coordinate complex, multi-step logic. It employs an onion-model middleware approach for cross-cutting concer

ZenML is an extensible machine learning orchestration framework designed to manage the end-to-end lifecycle of data pipelines and AI agent workflows. It functions as a durable orchestrator that executes machine learning tasks as directed acyclic graphs, ensuring that every step is containerized for consistent performance across local, cloud, and hybrid infrastructure. By decoupling pipeline code from underlying compute and storage backends, the platform allows developers to define infrastructure-agnostic stacks that remain portable across diverse environments. The project distinguishes itself