Packages

Samba server provides file and print services for various clients, allowing for interoperability between Unix/Linux and Windows systems.

Open-source context database for AI agents that unifies memory, resources, and skills management through a virtual filesystem paradigm with hierarchical three-tier context delivery (L0/L1/L2) and self-evolving capabilities.

NGINX Modern Application Reference Architecture (MARA) providing foundational examples for deploying modern applications using Pulumi across multiple cloud providers with NGINX Ingress Controller. Archived and unsupported.

A Blender addon that enables rapid snapping of objects, vertices, and points to various targets (origins, vertices, curve points, edge midpoints, face centers), replicating Maya/3ds Max snapping workflows inside Blender.

Multi-agent orchestration toolkit that automatically discovers, creates, and evolves AI agents within governance boundaries using MongoDB-backed persistence.

DNS proxy for penetration testers and malware analysts. Intercepts DNS queries and returns forged responses, enabling traffic redirection for security testing. Supports selective domain filtering, wildcards, multiple record types (A, AAAA, MX, CNAME, NS, PTR, SOA), IPv6, TCP/UDP, and INI-based configuration files. Cross-platform (Linux, Windows, macOS, iOS, Android).

Dense Extreme Inception Network (DexiNed) for edge detection — a PyTorch deep learning model that detects edges in images via single-stage training from scratch, without requiring pre-trained weights. Achieves 0.895 ODS on BIPED dataset.

A command-line penetration testing tool for accessing and controlling remote Windows systems using methods like WMIEXEC, PSEXEC, SMBEXEC, and ATEXEC with credential or hash-based authentication.

A curated list of production-ready plugins for Claude Code, enabling custom commands, specialized agents, hooks, and integrations across 500+ applications via the Composio platform.

Sage Mode for F# development — REPL with solution or project loading, Live Testing for FREE, Hot Reload, and session management.

Sync skills from Git repositories to local AI coding tools.

Automatic Exploit Generation (AEG) tool that uses symbolic execution (angr) to find and weaponize vulnerabilities in CTF binaries, supporting buffer overflows, format strings, shellcode injection, and ROP chains.

Converts Git repository contents into a text format suitable for feeding into AI language models, preserving file structure and contents for code review and documentation generation.

Centralized operations and governance dashboard for managing AI agents via OpenClaw Gateway, with task assignment, approval workflows, audit trails, and multi-team orchestration.

A collection of tools for browsing and searching archived Google Reader data after the 2013 shutdown, including a zombie_reader that recreates the original Reader interface and a command-line item_lookup tool for searching specific items.

Free, open-source Human Capital Management platform built on SugarCRM/SuiteCRM foundations providing recruitment, time tracking, onboarding/offboarding, leave management, resource booking, travel/expense tracking, performance evaluations, and mobile apps.

Collection of OpenClaw extension plugins enabling AI integration with Chinese IM platforms (DingTalk, QQ, WeChat Work, Feishu). Supports text, markdown, images, files, voice messages, streaming responses, and scheduled tasks.

An OpenClaw plugin that intelligently routes LLM requests to the most cost-effective model using a 23-dimension scoring algorithm, potentially reducing costs up to 70%. Includes real-time dashboard, usage tracking, alerts, and OTLP observability support.

Open-source, privacy-first AI meeting assistant and Cluely alternative. Provides real-time speech-to-text transcription across Zoom, Teams, and Google Meet with AI-powered contextual assistance, local RAG for conversation memory, screenshot analysis, and a stealth mode that is invisible in screen shares.

Multi-agent AI coding system where an orchestrator agent delegates to specialized explorer and coder subagents with intelligent context sharing. Ranked 13th on Stanford's TerminalBench leaderboard, outperforming Claude Code on that benchmark.