Claude
Tools

Agent profile manager for Claude Code & Codex. Per-directory profiles select which skills, MCP servers, and plugins load — automatically, before launch. Install: npm install -g cue-ai

Portable AI-workflow control plane for Claude Code — one install.sh ships specialized agents, user-invocable skills, always-on rules, lint/audit/secret hooks, the 6-gate review, design templates + sprint discipline, an /onboard wizard, and a pull-based feedback loop. v0.4.0

L'humanizer italiano con i superpoteri: via i segni dell'AI, e testi anche corretti e ben scritti.

Agent skill that scaffolds Alchemy (Infrastructure-as-TypeScript) into any codebase with strict secret hygiene. SKILL.md format, skills.sh + npx compatible.

Cursor / Claude Code skill pack for schwab-marketdata-mcp (60+ refs, 4 playbooks, EN+zh-CN bilingual).

Turn messy ideas into agent-ready prompts. A small skill for Claude Code, Cursor, built on Anthropic’s prompt-engineering principles.

An opinionated, fully-worked Claude Code configuration: path-scoped rules, skills, subagents, deterministic hooks, and a worktree-first workflow. Demonstrated on a TypeScript monorepo; the framework adapts to any language or stack.

LLM Wiki skill for Claude Code — Karpathy LLM Wiki pattern, batteries included

Claude API CLI demonstrating all 5 CCA-F certification domains — chat, tool use, agents, MCP, and safety guardrails in a single production-pattern tool.

The missing architectural layer above MCP. Production patterns for LLM agents wired to enterprise systems at scale — without context blowout, hallucinated tool selection, or audit gaps.

Coordinate multiple Claude Code terminals on one project — file locks, shared kanban, planner approval gate. Battle-tested in production.

Local observability for AI agent sessions — token usage, cost estimates, summaries, and efficiency insights. VS Code extension and standalone Docker image for GitHub Copilot, Claude Code, and Codex.

Drive your own already-open browser from your AI coding agent (any MCP client) via a self-hosted relay + MV3 extension. Tools: screenshot (+full-page), navigate, click, fill, query, new_tab, evaluate (arbitrary JS, even on CSP-locked sites via userScripts). Toolbar activity flash + sqlite audit log for visibility.

Claude Code skills to help you think like a Senior Engineer.

Claude Code TTS and voice hook setup for WSL2 — hooks, scripts, and config for speech input and spoken responses.

Idea to verified, working code through five single-responsibility AI-agent skills. The spec is the single source of truth.

Hands-free voice output for Claude Code — reads AI responses aloud via Microsoft Edge neural TTS

A Claude Code skill for generating Anthropic-style standalone HTML diagram files.

苏格拉底对话体写作 Claude Skill — 一句『把 X 写成苏格拉底对话』→ 信徒陈词 + 苏格拉底层层追问 + 反转 + aporia 收尾,七条铁律(信徒有牙/苏格拉底不认错/不许独白/结论 reframe/去 AI 腔)+ 柏拉图真功夫(逼定义/反例击穿/先骗同意再合拢),可选网页排版模板

A small, deterministic safety layer for Claude Code — five PreToolUse hooks between tool calls and your filesystem.

Agent skills for publishing engineering docs, repo summaries, PR digests, changelogs, API docs, onboarding guides, and specs to Uselink.

Bitbucket Cloud CLI built for AI coding agents — single-file bash with zero-config repo auto-detect and inline PR review, plus CLAUDE.md snippet, Rule, and Skill bundled out of the box.

Plumbline — a self-learning, customer-value-governed agile AI agent team for Claude Code. 87 subagents + skills, TDD defense-in-depth gates, Kaizen retros, a four-body adversarial council, and an empirically benchmarked QA harness. Does it hang true?

Principal-engineer operating mode for Claude Code: 5-phase loop, 30 red flags, cognitive layers for software and security, grounded chapter-by-chapter in 10 canonical engineering books. MIT.