paperetl is a Python ETL library that normalizes PDFs, PubMed, arXiv, TEI, and CSV metadata into a unified article schema, supporting SQLite, JSON, YAML, and Elasticsearch storage.
Monopoly-core is a Python library and CLI for converting bank statement PDFs to CSV using per-bank parser classes. It supports 20+ banks, OCR, and safety checks.
PartCrafter generates multiple semantically distinct 3D mesh parts from a single RGB image using latent diffusion transformers, enabling structured 3D generation with pretrained models and VLM-based part suggestions.
pdf-document-layout-analysis is a Dockerized microservice using Vision Grid Transformer and LightGBM for PDF layout analysis, offering high accuracy or fast processing with OCR, translation, and multi-format export.
pentest-agents deploys 50 specialized AI agents across 7 coding tools with a multi-IDE portability layer, autonomous exploit chains, endpoint brain, and MCP servers for bug bounty hunting.
Pixal3D generates high-fidelity 3D assets with PBR textures from a single image using pixel-aligned projection conditioning. It offers a three-stage cascade and low-VRAM mode for consumer GPUs.
ProxmoxMCP-Plus is a Python-based toolset enabling advanced interaction with Proxmox VE via MCP protocol. It offers flexible config and runtime modes for sysadmins and devs.
QSTrader offers a modular Python backtesting framework for long-short equity strategies using daily OHLC data and calendar-driven rebalancing. Its clean separation of signal, portfolio, and execution components stands out.
ReasoningBank introduces memory-aware test-time scaling for AI agents by storing reasoning traces from both successes and failures, enabling self-evolution through experience.
SafestClaw uses classical ML pipelines and local AI models to deliver 90% of OpenClaw’s features at zero cost, avoiding prompt injection and cloud dependencies.
SAM3-UNet adapts Meta’s SAM3 foundation model for dense prediction tasks using a parameter-efficient adapter and U-Net decoder, enabling training under 6 GB GPU memory.
scenario-lab is a Python-based tool for running scenario simulations via a CLI, emphasizing reproducible workflows and modular structure with Python 3.12 venv support.
SceneSmith uses GPT-5-powered agents to generate physically plausible 3D indoor scenes from text prompts, ready for robotics simulation without manual cleanup.
Seeker hosts fake web pages to trick users into granting browser location permission, harvesting precise GPS and device fingerprint data via HTML5 APIs. Built with Python and Flask, it runs on multiple platforms and supports export to Google Earth and Telegram.
Skill Conductor enforces design patterns and uses a 5-mode lifecycle to manage AI agent skills, avoiding common pitfalls like the ‘description trap’ for more reliable skill development.
Spotify2YoutubeMusic is a Python app that migrates playlists and liked songs between Spotify and YouTube Music using smart caching and batch processing for efficiency.
SuperClaude transforms Claude Code into a structured AI development platform using behavioral instruction injection, 30 slash commands, 20 specialized agents, and 8 MCP server integrations for faster, token-efficient workflows.
Supertonic-3 is a Python TTS library running fully on-device via ONNX runtime, supporting 31 languages, zero-shot voice cloning, and a drop-in OpenAI-compatible API for local TTS deployment.
SupoClip is an open-source self-hosted AI video clipper using AssemblyAI transcription and multiple LLM backends including local Ollama. It runs on Docker Compose with FastAPI and Next.js.
SVFR combines blind face restoration, colorization, and inpainting in a single stable video diffusion model, enabling efficient multi-task video face enhancement.
