Browse agent skills

FAUXCLAW - Real Claude. Fake Bill. Free AI proxy for Claude Code

AI-powered codebase analyzer — scans project structure, builds dependency graphs, generates LLM-optimized prompts with minification, skeleton mode, and JSON patching

Elasticsearch bulk transport for loggerjs.

Tenant SDK for the AuspexAI volunteer compute network

An autonomous self-healing CLI manager that intercepts terminal command failures and automatically diagnoses and patches them using LLMs.

Core agent framework — base class, lifecycle, handlers, config loader

AI-driven browser automation — Stagehand wrapper with vault integration and session recording

Lightweight plan mode for Pi coding agent

HWP/HWPX to PDF converter with Hancom-fidelity layout, OFL font subsetting, ToUnicode-correct text layer (search/copy/extract), and invisible text-layer completion.

Fast C-backed trigram language model for word prediction and sentence completion

No description available

Zero-config LLM cost & quality monitor for indie AI builders. One-line SDK, local dashboard, budget kill-switch.

Session memory & orchestration for Claude Code: one index of all your sessions, three modes — finish / handoff / spawn

Agent skill for Cursor, Claude Code, and Codex — SEO, GEO, and AI visibility for Next.js and Vite/React apps. Covers metadata, Open Graph, Schema.org JSON-LD, sitemap, robots.txt, keyword clustering, llms.txt, and CSR/prerender guidance.

Skill pack para criar e auditar websites modernos com Next.js + TypeScript + Tailwind + shadcn/ui

GPU Flow CLI — European LLM inference, sandboxes and dedicated GPU.

Prompt lookup tool. Agent names a regression signature from a fixed list; bratch returns the prompt for that regression signature. The prompt tells the agent how to check the diff against a baseline for that regression.

n8n community node for the Apify Google Maps Email Extractor - scrape Google Maps businesses with emails, phones and social links for lead generation. No API key required.

Interactive skill visibility manager for pi coding agent — browse, toggle, and invoke skills from a keyboard-driven TUI overlay.

Connect your users' AI accounts to your app. Capped, revocable App keys, never raw credentials. Outlet is never in the data path.

让 AI 写出"你导师改过三遍之后"的那种论文 ✍️ 中文核心期刊文体级写作 Skill

Live DAG of Claude Code agents — watch parallel subagents fork, call tools, and return on one calm canvas.

Live deck of Claude Code and Codex agents — watch parallel subagents fork, call tools, and return on one calm canvas. Also available as npx ccdeck and npx agent-dag.

Live deck of Claude Code and Codex agents — watch tool calls, token spend and every Claude Code subagent on one calm canvas. Run it with npx ccdeck.

Prompt lookup tool. Agent names a vulnerability pattern from a fixed list; bspector returns the prompt for that vulnerability pattern. The prompt tells the agent how to check the skill manifest for that vulnerability.

This is a package holder for Gaunt Sloth, for future use, Gaunt Sloth is currently available as (https://www.npmjs.com/package/gaunt-sloth-assistant)[https://www.npmjs.com/package/gaunt-sloth-assistant]

MiMo 工作流技能包 — TDD、系统化调试、方案设计等 15 个技能 + 3 个 agent（mimo-compose / mimo-dev / mimo-review），适配 OpenCode

Prompt lookup tool. Agent names a smell category from a fixed list; bsmell returns the prompt for that smell category. The prompt tells the agent how to check the session for that smell.

Prompt lookup tool. Agent names a task class from a fixed list; banchor returns the prompt for that task class. The prompt tells the agent how to check the task against a stated mission.

Evidence-first design review for Apple apps. Capture the screen, read the AX tree, fix the slop.