GitHub 2026-04-01 Weekly Trending Projects

Last30Days is a multi-source trend briefing tool that searches and scores signals from social platforms, news, and prediction markets over the last 30 days, then synthesizes cited narratives—designed for researchers, prompt engineers, and product teams tracking short-term trends.

VibeVoice is Microsoft's open-source voice synthesis framework that combines low-frame-rate continuous speech tokenizers, an LLM, and diffusion generation to target long-form multi-speaker and low-latency realtime scenarios; suitable for research and prototyping but carrying compliance, licensing, and misuse risks.

For creators and researchers, Deep-Live-Cam enables real-time face-swapping and one-click video deepfakes from a single image with GPU and Apple Silicon acceleration; useful for creative demos and live shows but requires strict legal and ethical compliance.

Extensible super-agent runtime with skills, sandbox execution and memory.

AI Scientist-v2 offers an end-to-end agentic pipeline to ideate, run experiments, and draft papers; it requires sandboxed GPU environments and has unclear license and maintenance guarantees.

oh-my-claudecode delivers zero-config, multi-agent CLI orchestration for the Claude Code ecosystem—team pipelines, parallel execution and cost optimization—however its license and maintenance status require verification.

Pascal Editor offers node-based 3D building editing with React Three Fiber and WebGPU for real-time rendering and system-driven geometry generation, suited for teams building interactive modeling and visualization prototypes; evaluate maintenance and licensing risks before production use.

Hermes Agent is a server-deployable autonomous AI agent offering a terminal-first interface, multi-platform messaging gateway, persistent memory and skills hub, sandboxed terminals, and scheduling—suitable for long-running, auditable, and extensible automation in engineering and operations.

Superpowers provides a composable skills library and TDD-enforced agent workflow for Claude Code, enabling end-to-end automation from design and planning to subagent execution, with plugin integration and extensibility.

Everything Claude Code delivers a production-grade toolchain for agent harness performance and governance—covering skills, hooks, memory persistence, and security scanning—suited for teams requiring cross-platform parity and measurable quality gates.

TradingAgents constructs a modular, configurable research pipeline of role-based LLM agents, supporting multi-provider models and simulated execution—suited for academic and R&D strategy validation but not investment advice.

A Letta-backed background agent that watches Claude Code sessions, indexes code, and injects persistent memory and guidance for session continuity.

Offline-first server with local AI, offline Wikipedia and education content.

MiniMind is a from‑scratch micro‑GPT toolkit enabling rapid single‑GPU training and hands‑on learning of pretraining, fine‑tuning and distillation workflows.