Search

A Claude Code skill that turns Claude into a design reviewer that SEES your page. Uses Playwright MCP to capture desktop/tablet/mobile screenshots, audit typography, contrast, rhythm, hierarchy — and propose concrete CSS fixes. Bilingual EN/RU.

PRISM — A multi-LLM peer review framework for architectural governance. Triangulate AI Architecture Decisions.

Claude Code Manager — manage Claude Code projects, sessions, and memory

Design methodology reproduction kit from 小耳 Art Studio — Japanese-editorial aesthetic portable across Claude Code / Hermes / openclaw / any LLM. CC-BY-4.0.

Turn lecture videos into Markdown documents pairing each slide with cleaned narration. Four-stage local pipeline + bundled Claude Code skill.

A Claude skill for humanities and social-science academic writing — thinking partner, not polishing tool. Four-layer critique, anti-sycophancy devil's advocate, voice preservation, reflexive-writing support, ADHD-friendly.

Claude Code skill: orchestrate parallel teams of Claude agents — breadth/depth/mixed strategies, task coordination, optional per-agent TTS

Claude Code skill to stop it from burying the answer. ADHD-friendly output.

A precision-engineered AI workforce for fiduciary-grade financial strategy. Modeled after the NeuraForge architecture, FinForge AI uses a Unified MCP Hub and the FinWorth Math Engine to provide zero-hallucination tax optimization, debt reduction roadmaps, and portfolio auditing. Compatible with Cursor, Claude Code, Kiro CLI, and ChatGPT.

Pydantic data models and adapters for the LLM Protocol Suite.

QwenPaw LLMVTuber Channel - WebSocket频道，对接Open-LLM-VTuber前端

Black-box safety eval infrastructure for Indic LLMs

Microsoft Azure Storagesync Management Client Library for Python

ZBAL pronounced [zi:bəl], stands for Zero Buffer Acceleration Library. It contains a bunch of well tuned operators for LLM inference and training, which has two key advantages: zero intermediate buffer and blazing fast.

Boruna: a deterministic, capability-safe programming language and framework for LLM-native applications