Claude
Tools

MCP server for Sellercloud REST API and legacy web — 73 tools for Claude Desktop and other MCP clients

AI-powered options trading assistant for Interactive Brokers. Real-time Greeks, McMillan/Overby strategy library, wheel tracking, P&L analytics, earnings warnings. Plug-and-play Claude Code skill.

Personal Claude Code hooks.

Publish AI-generated HTML to your own Vercel — a Claude Code skill + setup wizard.

PPTWork has two portable Anthropic-style Skills for AI agents to plan, author, and export PowerPoint decks. Both work in OpenCode, Claude Code, Cursor, Codex, or any other host that honors the SKILL.md frontmatter convention.

Free, vendor-neutral skill that turns occupancy and presence data into a decision-ready report. Works with Claude, ChatGPT, Gemini, and other LLMs.

A method-driven computational biology PI persona for Claude Code — stops asking, starts deciding. Targets Nature sub-journals.

A peer-review protocol for AI models. Claude, Codex, Gemini and others review each other's code via CLI handoffs. Bidirectional, symmetric, opensource.

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.

Async Python implementation of the Claude Code agent harness — real Bash/Read/Glob/Grep tools, three-tier caching, multi-stage compaction, allow/ask/deny permissions, Pre/PostToolUse hooks, slash commands, subagent isolation. OpenAI + Anthropic + Ollama. ~1.9K LOC, 14 commits.

My Claude Code statusline + supporting hooks (model, cost, context, git, worktree, skills).

A Claude Skill for extracting personal principles from your own life experiences, inspired by Benjamin Franklin's 13 virtues method.

LLM-driven automated penetration testing pipeline — subdomain → probing → endpoints → fuzz → JS audit → auth analysis → API scoring → unauth API verification. Evidence-backed, multi-format reports, Tool Cards, Planner/Verifier/Reporter contracts.

Spec-Driven Development as a single Agent Skill

Local observability for Claude Code. Captures sessions, sub-agents, and tool calls to Postgres via HTTP hooks.

自主用デモ

Spec-driven workflow plugin for Claude Code / CodeBuddy with advisory hooks, phase selectors, locks, and task-swarm.

Claude Code hooks for GeekMagic SmallTV-Ultra display

Safety guardrails for Claude Code running unattended. Hooks that block dangerous tool calls before they execute, feed structured reasons back to Claude so it self-recovers, and write an audit log.

Operational Tooling. Better token consumption. Faster decision making.

Solo-dev Claude Code skills: brainstorm → plan → code. No ceremony, no vibes. Fusion of Matt Pocock, OpenSpec, get-shit-done.

MCP server giving Claude and other LLM clients structured access to the Internet Archive's Wayback Machine — 6 tools, 4 prompts, 1 resource, over stdio.