Headlong is a Bash‑based micro‑harness that runs a continuously thinking LLM‑backed agent. The core `shellm` engine lets the model write Bash commands, executes them, and feeds the output back as context. Agents persistently generate thoughts, can be accessed via Slack, Telegram, or a web dashboard, and store their entire history as a JSON‑L DAG with tiered summarisation. Installation is a one‑liner that sets up Docker sandboxing (or host mode) and a unified `llm` CLI for Anthropic, OpenAI, Gemini, OpenRouter, or any local OpenAI‑compatible server (Ollama, llama.cpp, vLLM, etc.). The framework emphasizes tiny, composable Bash tools, self‑improvement via fork‑test‑merge, and safety via sandboxing and spend‑capped API keys.

A series of unified multimodal models for 3D world modeling that enable camera-centric spatial intelligence and 3D world generation using native physics, geometry, and appearance states.

A set of AI-driven workflows and curated lists for managing SEO content production and high-quality backlink submissions to product directories.

An open-source desktop application built with Tauri and React that provides a graphical interface for working with AI agents via the Goose backend.

A private, open-source Android keyboard that integrates cloud and on-device AI for proofreading, translation, Whisper voice typing, and OCR text extraction.

A Claude skill that enables the AI to watch videos from URLs or local files by extracting key frames and transcripts for multimodal analysis.

A local, swarm-intelligence global forecasting system that fuses over 40 real-time data feeds into a 3D intelligence globe to predict world events.

A self-hosted web platform for running and managing multiple virtual iPhones in a browser using Apple's Virtualization.framework.

An end-to-end solution for creating digital avatars by fine-tuning LLMs on personal chat history, featuring built-in privacy filtering and multi-platform deployment.

A unified AI model gateway and governance infrastructure that centralizes model access, cost management, and operational monitoring for AI applications and agents.

A set of npm‑based plug‑ins that extend the terminal‑based coding assistant *Pi* with UI footers, session search, auto‑permissions, sub‑agents, worktree handling and other utilities.

Pentest Swarm AI is an open‑source Go tool that runs dozens of autonomous LLM‑driven agents in parallel, coordinated via a shared Postgres blackboard with pheromone‑based stigmergy. It supports any LLM provider (Claude, Together AI, OpenAI‑compatible, Ollama, etc.), bundles a recon/exploit toolchain (ProjectDiscovery libs, nmap), and offers both an interactive terminal UI and a web dashboard. The swarm can target real hosts or bundled vulnerable labs, enforces scope, and produces evidence‑backed reports. Core features are stable; the swarm scheduler is in alpha, with further adapters and a fine‑tuned model planned.

A family of Conformer-based acoustic models for state-of-the-art speech recognition and emotion detection, with strong support for Russian and 70+ other languages.

An open-source MySQL branch that integrates DuckDB for high-performance analytics and native HNSW vector search for AI applications.

A community-driven aggregator that provides a unified, OpenAI-compatible API and GUI to access multiple LLMs and media-generation models from various providers.

Typesense is an open-source, fast, typo-tolerant search engine built in C++ that serves as a simpler alternative to Elasticsearch and a self-hostable alternative to Algolia. It supports full-text, vector, semantic/hybrid, and conversational (RAG) search, plus image and voice search, all through a simple HTTP API.

A dynamic, high-performance API gateway that provides traffic management and AI proxying capabilities to route and secure traffic between services and LLM providers.

A Python research project that compares four modern time‑series models (PatchTST, TFT, N‑HiTS, CatBoost) for hour‑by‑hour electricity‑load forecasting using real Russian utility data. Includes data preprocessing, training pipelines, statistical analysis, and visualisation, and is released under Apache 2.0.

A ComfyUI node pack for instruction-based image editing using Krea 2, enabling high-fidelity identity preservation through dual latent and semantic conditioning.

Timeline Studio is a browser‑only, local‑first AI video editor that runs on WebGPU/ONNX. It offers multilingual TTS, AI‑generated music, automatic captions, smart framing, watermark removal, vocal separation, and digital‑human avatars, all without uploading media. Projects are saved as portable `.timeline` files and can be edited via a rich UI or a head‑less CLI skill that lets LLM agents inspect, plan, and apply edits programmatically. The code is MIT‑licensed; model weights are mirrored on Hugging Face and ModelScope under their original licenses.

A self-hosted control plane for operating AI agents, allowing users to dispatch tasks, track spend, and coordinate multiple agent runtimes from a single dashboard.

A lightweight, terminal-first AI-native development environment that integrates a code editor, GPU-accelerated terminal, and agentic AI assistants into a single workspace.

An agent skill that transforms complex academic papers into structured, evidence-based Obsidian notes, automating the ingestion layer for academic wikis.

An autonomous agent framework for Elixir that provides a formalized architecture for state management, multi-agent orchestration, and fault-tolerant workflows.

roborev is a Go‑based tool that adds a Git post‑commit hook to run AI‑powered code reviews locally. It integrates with popular coding agents (Claude Code, Codex, Gemini, Copilot, etc.), shows findings in a terminal or browser UI, and can automatically fix issues via an auto‑fix/refine loop. All processing runs on the developer’s machine, with exportable analytics and extensible hooks.

A collection of 164 specialized skills for Claude Code that guide developers through the entire process of building, launching, and growing apps for iOS, macOS, watchOS, and visionOS.

A high-performance neural network inference framework optimized for mobile, embedded, and desktop deployment with no third-party runtime dependencies.

An open-source platform for tracking and optimizing AI visibility (AEO/GEO), allowing brands to monitor how they are mentioned and cited by AI answer engines.

A collection of standardized skills for AI agents to generate professional diagrams, data visualizations, and technical documentation in Markdown.

A low-level orchestration framework for building controllable, stateful AI agents with support for durable execution, long-term memory, and human-in-the-loop oversight.

Statewave is an open‑source, self‑hosted memory runtime for LLM agents. It records raw events (episodes), compiles them into typed memories (via heuristic or LLM compilers), and serves deterministic, token‑bounded context bundles with full provenance. Built on PostgreSQL + pgvector, it offers policy‑driven labeling, multi‑tenant isolation, and SDKs for Python and TypeScript.

A portable operating harness and Python API that enables AI agents to control Android and iOS devices locally or in the cloud.

An open-source platform for organizations to manage, secure, and govern their AI ecosystems, providing centralized control over LLM access, MCP servers, and AI activity auditing.

An offline-first personal diary app with rich media support, biometric security, and an AI assistant for mood suggestions and chat.

A production inference stack that decouples AI workloads from proprietary hardware, allowing a single codebase to run models on NVIDIA, AMD, Intel, and TPU/Trainium accelerators.

A streaming-optimized Markdown renderer for React that handles incomplete AI-generated content gracefully to prevent formatting glitches.

A local-first AI workbench for creating, animating, and exporting 2D game art, featuring identity-preserving AI animation and a deterministic Rust-based rigging engine.

A browser automation tool that allows AI coding agents to maintain persistent browser sessions and use compact accessibility trees for efficient web interaction.

Sessions Viewer is a cross‑platform desktop app (Tauri + Vue 3) that lets you open, search, replay, and continue local transcript files from AI coding assistants like Claude Code, Codex, Grok Build, Kimi Code, Pi, Antigravity CLI and opencode. It provides faithful replay of tool calls, global search, split‑pane layouts, token‑cost analytics, and one‑click resume of sessions in embedded or external terminals, all while keeping the original JSONL files read‑only.

A structured prompt generation tool for AI painting that turns complex prompts into a visual fill-in-the-blank interaction for easier management and iteration.

An MCP server that integrates the SearXNG API to provide AI assistants with privacy-respecting web search and URL content reading capabilities without requiring paid API keys.

MLC LLM is a universal deployment engine and ML compiler that enables high-performance native execution of large language models across diverse hardware and platforms.

A family of multi-platform GUI agents and foundation models that automate tasks across mobile, desktop, and web interfaces using visual perception and planning.

CORAL is an open-source infrastructure for autonomous AI agent organizations to perform autoresearch by running experiments and sharing knowledge to evolve solutions.

A comprehensive framework for building and deploying autonomous AI agents, featuring a five-layer stack for managing prompts, context, tools, and execution loops.

An open-source infrastructure layer for building and operating persistent, coordinated, multi-agent systems at organizational scale.

A local knowledge base QA system that uses two-stage retrieval to provide accurate answers from various file formats while supporting fully offline deployment.

A behavioral audit engine for fund managers that analyzes years of trading data to determine if returns were driven by skill or market luck, producing detailed interactive reports.