An open-source app builder that generates applications from user intent and uses a separate verification harness to ensure every feature meets business requirements.

An open protocol for AI Agent work contracts and verifiable execution, ensuring that agent actions are authorized, evidenced, and formally accepted by humans.

A persistent, local-first memory store for AI agents that provides durable, typed, and searchable memory using a single Rust binary without requiring a hosted vector database.

Session Kit is a local command‑line dashboard that lists all your Claude Code and Codex AI coding sessions, shows their state (e.g., waiting for input), and lets you jump to any session by typing its stable number. Sessions survive terminal or SSH disconnects, can use different provider accounts, and are protected by a safety model that re‑verifies the exact provider process before acting. It installs via a verified release, runs on Linux (systemd) or macOS 14+, and stores all data locally with no telemetry.

A set of experimental agent presets for DeepSeek that anchor the model's initial reasoning trajectory to a high-performance minimal state before promoting it to a full toolset.

A management hub for the NullClaw ecosystem that simplifies the installation, configuration, and monitoring of AI agent components through a unified web UI and CLI.

A browser-based robot teleoperation platform that crowdsources human demonstration data for training Vision-Language-Action (VLA) models and imitation learning policies.

A self-hosted AI fashion platform that digitizes your wardrobe and uses generative AI to recommend and visualize outfits based on weather, occasion, and style.

An agentic QA harness that allows users to write web and mobile tests in natural language and uses self-healing AI agents to adapt to UI changes.

An AI-powered formal verification tool that automates the creation of specifications and proofs to provide machine-checkable evidence of software correctness.

A reference layer for multi-agent systems that replaces bulky context handoffs with 30-byte tokens, providing verifiable telemetry and tailored data projections.

A Python package that converts documents and scanned images into clean, structured markdown chunks using OCR and LLM-powered semantic chunking for RAG systems.

A collection of ready-to-deploy AI pipeline templates for high-accuracy, real-time RAG and enterprise search that syncs directly with live data sources.

An AIOS for real-time multimodal applications and smart hardware, providing an integrated development experience from AI-native coding to on-device inference.

A human-centric framework for assessing the stealthiness of adversarial perturbations in images, featuring a new dataset (APSD) and an attention-based prediction model (A2SM).

A command-line interface for unified multimodal AI generation, allowing media teams and AI agents to generate images, videos, audio, and text through a single API key and credit balance.

A swarm of disposable AI agents that interact with the web using unique, temporary identities to prevent tracking and fingerprinting.

A minimal, MIT‑licensed *skill* that lets AI agents obtain disposable phone numbers and read SMS verification codes via a commercial API, enabling automated sign‑ups and 2FA without human interaction.

This is a legitimate software project: an inference-time control suite (Python scripts, state files, protocols, and test suites) that AI agents load as a skill to manage complex engineering tasks. It imposes discipline — source re-reading, semantic repository maps, evidence tracking, verification gates, and multi-agent coordination — so agents produce grounded, verifiable results. It works through instructions and external state rather than modifying model weights.

Agentlas OS is an open‑source, cross‑platform framework (Hephaestus engine) that lets you build, borrow, and own portable LLM‑driven agents or teams. Agents are packaged method documents with strict JSON contracts, stored in a private “Agent Cloud” or fetched from a public Hub, and can run on any supported host (macOS, Windows, Linux) with any LLM you already use. Installation is a single script; the command surface (`/agentlas build`, `/agentlas hub`, `/agentlas upload`, etc.) and a visual Desktop UI let you compose, verify, and execute agents while keeping credentials and permissions local.

A portable agent skill that combines conversion-focused copywriting with a 33-point humanizing engine to remove AI-generated markers and create high-converting marketing text.

floe‑guard is a Python/TypeScript library that tracks the real USD cost of every LLM, speech‑to‑text, text‑to‑speech, and other AI‑vendor call in‑process, and can hard‑stop a call before it would exceed a user‑defined dollar ceiling. It ships with adapters for OpenAI, Anthropic, Gemini, LangChain, CrewAI, LiteLLM, Vercel AI SDK, and voice platforms (Pipecat, LiveKit, Vapi, Retell). A bundled public‑price cost map provides offline pricing; you can override it with your own rate card. Budgets can be persisted across processes via SQLite, and an optional hosted Floe service adds a Coverage Score and 7‑day history. The core API is `BudgetGuard(limit_usd)`, with `check()` (pre‑call) and `record()` (post‑call) methods, plus atomic `reserve_tool`/`settle_tool` for paid tools.

A pipeline for turning images into 3D meshes that allows users to run and compare multiple open-source reconstruction models to find the best output.

Hezo is an open‑source server + web UI that lets you build and run organised teams of AI agents (CEO, Captains, engineers, designers, etc.) inside sandboxed containers. You supply your own LLM provider keys, set token and container‑hour budgets, and the platform handles secret protection, Git integration, and a unified “meta‑harness” that normalises different model runtimes. Install with a single binary, self‑host or use Hezo Cloud, and manage projects via an org‑chart‑style UI.

A security protocol and enforcement gate that ensures autonomous AI agents operate within a finite, verified mandate to prevent unauthorized consequential actions.

Pathway is a Python‑first, Rust‑backed framework for building batch‑or‑stream data pipelines. It offers connectors to many sources, stateful operators, persistence, and a dedicated LLM toolkit (wrappers, parsers, vector index) that makes real‑time RAG and other LLM applications easy to write and deploy via Docker or Kubernetes.

A fully open, verification-first LLM-inference hardware tile that integrates KV-cache compression directly into the datapath to optimize edge inference.

A robotics-first observability and visualization stack for multimodal time-series data, providing synchronized 3D/2D viewing and Arrow-backed storage for robot sensor streams.

An optimized GPT-2 training implementation that uses memory and weight averaging techniques to set a new speed record for the modded-nanogpt challenge.

A scalable backend server that uses IBM's Docling to convert various document formats like PDF and DOCX into Markdown via a FastAPI and Celery-based API.

A collection of specialized agent skills for designing, building, and reviewing software in React, React Native, and security.

A task generator that creates new, distribution-matched ARC-AGI-1 style tasks to evaluate AI models on unseen problems and prevent benchmark contamination.

REDCELL is an open‑source, self‑hosted platform that uses LLM‑driven agents to run a full penetration test (real Kali tools, browser automation, reverse shells) and produce a polished report, all controlled through a live React console.

A provider-agnostic terminal AI coding agent that uses a Think-Act-Verify loop to read, edit, and execute code using any OpenAI-compatible LLM.

A trace-native CI/CD platform for AI agents that converts production failures into automated regression tests to block buggy releases.

An Indonesian-language crypto assistant that combines RAG for educational concepts with real-time, keyless market data to provide hallucination-free financial insights.

An open access and governance layer for AI agents that provides unified identity, policy enforcement, and auditable context management across an organization.

An autonomous research system that turns runnable, measurable projects into continuous, evidence-driven research runs to find optimal solutions through parallel agent exploration.

A long-term memory runtime for AI agents that uses Markdown files as the source of truth and a local index for ranked retrieval, enabling persistence across sessions.

A design-driven framework for generating professional, editable PowerPoint presentations that uses a visual verification loop (PPTX to PNG) to ensure high-quality delivery.

An open-source agent workflow for AI/ML research that automates idea generation and experiment execution to produce paper-ready evidence.

JiuwenSwarm is a multi-agent collaboration system that enables the orchestration of specialized agent swarms to automate complex tasks through natural language intent.

An automatic model router for Claude Code and OpenAI Codex that directs simple tasks to fast models and complex tasks to strong models based on per-turn analysis.

A browser automation skill for Codex that uses the Jev model to handle fast, low-cost navigation and clicks, while Codex manages high-level reasoning and verification.

A continual learning byte-level language model that can be trained from scratch on a single 8 GB VRAM GPU by paging weights from disk.

An ontology discovery and exploration platform that uses a disciplined epistemic loop to build a trustworthy, evidence-based knowledge graph for research.

A non-autoregressive decision model that provides calibrated, deterministic classification and routing in sub-25ms, replacing slow generative LLMs for high-speed operational tasks.

A collection of specialized decision-making skills and workflows for coding agents, enabling them to classify, score, and choose actions using the Jev system.