Claude Code agent skills for generating flawly architecture diagrams
Model Context Protocol (MCP) language server providing Kern UX component generation and strict accessibility validation.
Read-only investigation MCP server for AI agents to query obs-unified collector telemetry.
SpecLite local-first CLI control plane.
Local-first compatibility tester for OpenAI-compatible endpoints used by AI coding agents.
Unofficial local-first diagnostic CLI for NVIDIA NIM developer workflows.
Scan GitHub Actions workflows, agent instructions, and MCP configs for AI-agent injection and unsafe tool boundaries.
Convert Mermaid flowchart .mmd files into HyperFrames motion graphics and MP4-ready projects.
TypeScript SDK for the Meowtify API
PR call-graph tracer — extracts a route-rooted call graph from a PR diff and emits Mermaid + YAML artifacts via ts-morph static analysis.
AI-powered code to Mermaid flowchart — 代码一键生成流程图
featureReview — AI-powered route-level review for Next.js apps
Cross-platform installer that deploys a dual-system (Claude Code + opencode) AI-coding engineering rule system: global constitution, auto-triggering skills, and enforcement hooks.
hook vnode for Vue
Local-first security and cognitive-risk scanner for MCP tools and agent skills with ToolBOM, attack harness, and policy checks.
Zero-dependency diagramming engine for WorksheetJS — flowcharts, org charts, draw.io-style canvas.
Image extension for Typix
Policy-as-code admission controller for AI agent skills and MCP tools with SkillBOM, lockfiles, and supply-chain baselines.
Framework-agnostic sanitize and emit API for fintech provider errors (Plaid, Stripe, and similar).