Claude
Tools

Community binary patches for Claude Code CLI — unlock hard-coded limitations via Agent Skills

Generate production-ready cinematic video prompts for Higgsfield AI from reference images and project briefs.

Open-source SDK, CLI, MCP server, and Agent Skill for the Lacuna Music API — generate AI music from any Node runtime, shell, or AI agent.

Domain-specific AI skills for working with Valkey - reusable knowledge packs that help agents and developers write correct, idiomatic, high-quality code against the Valkey ecosystem.

🐦‍⬛ RavenClaude: Production-grade Claude Code setup for agent teams. One-click initialization of the Team Lead + Architect + Coder + Tester + Reviewer pattern with git worktrees, CLAUDE.md constitution, hooks, and strict quality gates. Level up from single Claude to a full AI engineering team.

Plugin para Claude Code com 40 skills de outbound B2B + briefing canônico que persiste entre sessões. Tunado para lemlist, agnóstico em fundo.

Discipline hooks for Claude Code. Stops the agent from claiming 'done' without proof, blocks accidental writes to ~/.claude config, and routes prompts to skills via local pgvector with zero Anthropic tokens.

Opinionated Claude Code skills for project bootstrap, persistent session memory, and multi-model code review (Claude + Codex).

WCAG 2.1 accessibility agent built on Anthropic API. Audits, evaluates and fixes digital products for UX, UI and development teams. Includes an installable Skill for Claude.ai and a full CLI with PDF/DOCX export.

Cross-agent skill that turns any research topic into a polished, interactive HTML field guide — sidebar nav, ⌘K search, SVG ecosystem map, 2×2 strategic matrix, 22-part editorial structure, VC/CB Insights depth. Claude / Codex / Gemini / Cursor.

Inventory dashboard for Claude Code — commands, skills, agents, hooks, MCP, plugins across user/project/local scopes. Zero LLM, zero servers.

Claude Code hooks that warn when a session starts burning tokens on repeated Reads, recursive greps, or WebFetch streaks

Claude Code hook that auto-converts Markdown to fancy infographic HTML on Write/Edit (with confirmation dialog)

Lazy-loading skill discovery for Claude Code via MCP - mirrors Anthropic's Tool Search Tool pattern for Skills. 100% local Apple Silicon stack (MLX) or cross-platform Docker.

ADAM — self-improvement layer for Claude Code. Observes session friction (corrections, retry loops, dead-ends, error loops, edit churn) via lightweight hooks and proposes targeted skills, memory entries, and agent edits. Auto-applies low-blast changes with cross-session evidence; queues everything else for review.

Multi-modal sharp thinker — FeynTech + Drama modes for Claude, Perplexity, ChatGPT, Manus

Minimal Go LLM client (v1.0, stable). Streaming, strict tool calling, adaptive extended thinking, prompt caching, token counting, cost helpers, retry middleware. Anthropic Claude, OpenAI (GPT-5 + Responses), Google Gemini, OpenAI-compatible hosts.

Audit sales offers (landing pages, pricing tiers, lead magnets) using Alex Hormozi's frameworks from $100M Offers and $100M Leads. Returns a scored audit with prioritized fixes. MIT licensed. By @johnericforte.

Claude skill that writes copy which sells. Print ads, OOH, billboards, headlines, taglines, landing pages, long-copy ads, emails, manifestos, product descriptions. Built on D&AD Copy Book craft, Bly direct-response frameworks, and Cannes Lions plus Clio Award print and OOH winners. Strict anti-AI writing layer. Claude Code + Cowork. MIT.

【Skill】AI 游戏实机视频 Prompt Pack 工厂

Sweep your ~/.claude/skills graveyard. Scans installed skills, MCP servers, and sub-agents, cross-references session logs, and shows what's actually used.

Nx-orchestrated fullstack monorepo: Astro + TanStack Start (web), NestJS + tRPC + Zod (API), Drizzle + Postgres (database). Fully integrated with Claude Code: custom agents, slash commands, hooks, and iterative Ralph loops are all pre-configured.

Agent Skill that teaches Cursor & Claude Code to output single-file HTML for spatial, parallel, comparable, or interactive content. 9 categories, 20 patterns, 12 ready-to-use templates. Inspired by Thariq Shihipar's 'unreasonable effectiveness of HTML'.

Substrate-based context compaction for Claude Code. Inspectable signal weights, local-only, MCP-ready. Alternative to /compact + hook-based capture (claude-mem, etc.).