发现优质的 AI Agent 技能

Build the ROSE source-to-source compiler in an isolated Docker container.

>

Connect Google Drive, Docs, Sheets, and Calendar to a Dockerized OpenClaw instance using **gog** (gogcli).

Connect Gmail to a Dockerized OpenClaw instance using **Himalaya** — a full-featured email CLI that supports reading, sending, and **downloading attachments**.

Install a fully isolated, production-ready OpenClaw instance inside Docker on macOS. One session, zero to running. All common pitfalls are handled inline.

>

> **CRITICAL:** Read `collect.md` completely before spawning any workers. > The #1 failure mode is writing the proof bundle before all workers report back. > This happened on midas-mcp (Feb 16, 2026). Never again.

You are now acting as the ClawSec Monitor assistant. The user has invoked `/clawsec` to manage, operate, or interpret their **ClawSec Monitor v3.0** — a transparent HTTP/HTTPS proxy that inspects all AI agent traffic in real time.

Unified web terminal for task management, queue processing, and system monitoring.

> **Power-user skill.** Sandboxer gives agents full access to tmux sessions, workspace files, and terminal output on your server. Intended for dedicated AI machines where agents run with root access. Not for shared or untrusted environments.

Tuna is a hybrid GPU inference orchestrator. It lets you deploy, serve, and manage LLM models (Llama, Qwen, Mistral, DeepSeek, Gemma, and any HuggingFace model) on serverless GPUs from **Modal, RunPod, Cerebrium, Google Cloud Run, Baseten, or Azure Container Apps**, with optional **spot instance fallback on AWS** via SkyPilot. Every deployment gets an **OpenAI-compatible `/v1/chat/completions` endpoint**.

Never lose a screen generation again. The only Stitch MCP proxy with **TCP drop recovery**.