菜单
Use when David wants to manage his VPS servers and the AI agents running inside them — connecting, deploying, monitoring, restarting, and operating remote hosts and their agents. Triggers on VPS, server management, remote host, SSH into server, manage my servers, agents on the server.
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Make an AI agent run on a schedule, loop, or interval — cron, heartbeats, recurring autonomous checks. Use for "run every N minutes", "schedule a task", "run on a loop", "heartbeat". Covers external clocks (Claude Code, Codex, Pi) vs Hermes' built-in scheduler.
MUST be read ANY time you interact with cmux in ANY way — listing/inspecting/creating/closing cmux workspaces, panes, or surfaces; reading or capturing pane/screen output; sending input or keys to a pane/surface; delegating to, polling, or checking on other agents running in cmux panes/surfaces; building or rearranging terminal layout; cmux browser automation; sending notifications/flashes/status/progress to the sidebar; editing cmux settings; or integrating an agent with cmux hooks. If your command starts with `cmux ` or touches a cmux workspace/pane/surface/agent, read this FIRST. Triggers on "cmux", "in this workspace", "this pane", "the other agent", "delegate to", "check on the agent", "send to the pane". macOS only (14.0+).
Explain and write effective instructions for OpenAI Codex's `/goal` feature — the persistent self-checking agent loop (plan → act → test → review → iterate). Use when the user mentions Codex `/goal`, "goal loop", "Ralph loop", wants to kick off a long-running autonomous Codex run, asks how to write a goal prompt, or wants a one-paragraph goal instruction drafted.
How to delegate work to another AI agent (Pi, Codex, Claude Code, Hermes) — picking the right agent, sending prompts to TUI agents, polling progress. Read BEFORE any `cmux send`/`tmux send-keys` to an agent, or whenever delegating, relaying, spawning, or orchestrating agent-to-agent work.
Score any AI model on the DeepSWE coding-agent benchmark via the OpenRouter API. Use when the user wants an independent, reproducible coding-agent eval — "run DeepSWE", "benchmark this model on DeepSWE", "score model X on the coding benchmark", "test a model via OpenRouter on DeepSWE", or to verify vendor-reported coding scores. Covers setup, the OpenRouter wiring for mini-swe-agent, single-task / subset / full 113-task runs, and leaderboard submission.
Keep David's MacBook awake with macOS caffeinate — prevent sleep, screen dimming, or both, for a set duration or while a process runs. Use when the user says "don't let my mac sleep", "keep the screen on", "anti-sleep", "caffeinate", or wants the machine awake overnight / during a long build.
| name | vps-server-management |
| description | Use when David wants to manage his VPS servers and the AI agents running inside them — connecting, deploying, monitoring, restarting, and operating remote hosts and their agents. Triggers on VPS, server management, remote host, SSH into server, manage my servers, agents on the server. |
Source of truth: <path-to-infrastructure-doc>/infrastructure.md (read it for the latest — IPs/expirations change).
| Hostname | IP | OS | Purpose | Expires |
|---|---|---|---|---|
| personal.openclaw | Ubuntu 24.04 (Dokploy) | OpenClaw — personal instance | ||
| n8n.automations | Ubuntu 24.04 (n8n) | All n8n workflow automations (primary) | ||
| private.hermes | Ubuntu 24.04 | Hermes Agent — Discord gateway |
SSH as root@<IP>.
<n8n-app-hostname>. Build/edit workflows, no server access. Safest to share.root@<IP>. Docker, files, system config. Trusted technical people only.hpanel.hostinger.com. Billing, reboot, OS reinstall. Exposes SSH creds + browser terminal, so it grants server access too. David only.For multi-step or exploratory work, SSH into the box first and launch the agent ON the VPS (e.g. codex --yolo), then talk to that local-on-server agent — it has full filesystem/process context and avoids fragile SSH round-trips. For short command sequences (update, config change, restart), driving an existing SSH session directly (e.g. via a cmux pane) is fine.
When checking on a remote/on-box agent, send David one concise status line each time: what it is doing and whether it is on track.
Claude Code cmux note: after Claude finishes, it may prefill a predicted next user message; that draft is Claude, not David speaking.
library/hermes/.hermes --version # shows version + commits behind
hermes update # auto-snapshots, updates deps, rebuilds web UI, restarts gateway itself
hermes gateway status|restart
journalctl --user -u hermes-gateway --since '5 min ago' --no-pager # gateway logs (systemd USER service)
~/.hermes/config.yaml under model.provider + model.default — NOT in .env. Change via hermes model (interactive) or edit the yaml directly, then hermes gateway restart to propagate to gateways.EBADENGINE warnings during update (deps want Node >=24, box runs v22) are non-blocking — do not "fix" them.library/hermes/.