Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add skills/weifeng-work/agent-node/skillnpx skills add weifeng-work/agent-node --skill skillgit clone --depth 1 https://github.com/weifeng-work/agent-nodeWhat it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00000 | $0.03115 |
| Opus 5 | $0.00000 | $0.01558 |
| Sonnet 5 | $0.00000 | $0.00623 |
| Haiku 4.5 | $0.00000 | $0.00312 |
Grade A, and why
skill scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured yesterday.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 239 lines — stays where its author put it; the contents beside it link to each section on GitHub.
agent-node 节点协作 SKILL 说明书
本文件是智能体（AI）使用 agent-node 节点能力的唯一官方说明书。 加载本 skill 即学会与本机节点及全网节点协作。
一、这是什么
agent-node 是局域网多机智能体协作节点。每台电脑跑一个节点，节点间自动发现、 点对点直连（无中心服务器）。你（AI）通过本机节点获得三样能力：
- 调用其他电脑上的智能体（执行器：CodeBuddy CLI/TUI、WorkBuddy 等）
- 跨机文件传输（推送/拉取/浏览任意目录）
- 跨机聊天与远程命令
安装（Windows）：PowerShell 一行
irm https://raw.githubusercontent.com/weifeng-work/agent-node/main/scripts/install.ps1 | iex——自动下源码 + venv 装依赖 + 建桌面快捷方式 + 启动节点。 安装会生成data/launch.json（启动器的外部启动清单，与内置回退链对齐）——由 install 维护，AI 不得手改。 日常控制用agent-node start|stop|status|restart|update， AI 能力调用用agent-node <子命令>。
二、接入方式（唯一入口）
前置条件：你能执行 shell 命令（PowerShell / Bash）。
安装
irm https://raw.githubusercontent.com/weifeng-work/agent-node/main/scripts/install.ps1 | iex
装完自动启动节点。刷新当前会话 PATH 让 agent-node 立即可用：
$env:Path += ";$env:LOCALAPPDATA\agent-node"
确认在线：
agent-node status
身份注册（首次，只需一次）
agent-node register
生成 ~/.config/agent-node/caller.json，内含你的 caller_id。
caller_id = 你的专属异步邮箱键——异步任务回执只进你的邮箱，其他 AI 看不到。
同一客户端稳定用同一身份文件，不要每次临时生成。
也可用环境变量 AGENT_NODE_CALLER_ID 覆盖（优先级高于配置文件）。
日常调用
所有操作统一走全局命令：
agent-node <子命令> [参数]
三、核心概念（防混淆）
| 标识 | 含义 |
|---|---|
node_id |
节点唯一身份（node-<主机名>-<短码>），寻址一律用它 |
name |
节点显示名（人改的，可重复，仅展示） |
agent_id |
执行器条目在本节点的标识（如 codebuddy-cli/workbuddy） |
executor_id |
全局寻址 = <node_id>/<agent_id>，AI 用它调执行器 |
caller_id |
你的身份（收件箱归属键，register 生成） |
四、子命令全表
本机
| 命令 | 说明 |
|---|---|
agent-node info |
本机节点概览（node_id/名称/team/端口/面板/三开关/在线） |
agent-node config |
读本机配置 |
agent-node rename <name> |
节点改名（仅本机） |
agent-node team <team_id> |
设置/切换 team（危险操作：断开旧 team 连接） |
agent-node switch <allow_shell|allow_file|allow_ai_task> <on|off> |
三开关 |
agent-node admin <on|off> |
管理员权限（需重启生效） |
重启节点使用生命周期命令：
agent-node restart（保持驻留，面板短暂中断后自动恢复）
节点
| 命令 | 说明 |
|---|---|
agent-node list |
在线节点列表 |
agent-node peer add <host> <port> |
手动加对端（跨网段兜底） |
agent-node peer remove <host> |
手动删对端 |
agent-node anchor list|add|remove |
锚点管理（被 AP 隔离时出站回连自愈） |
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- yesterday First seen · 239 lines · 0 tokens per session scan A 6d31f48ad48d
skill is a skill published in the GitHub repository weifeng-work/agent-node (0 stars, last pushed 10d ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 3,115 tokens. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
ha-merge-queue
Finds open Home Assistant pull requests that are genuinely ready to merge, checking CI, the merge-gate statuses, code-owner approval, merge conflicts, requested changes and unresolved review threads. Use when looking for PRs to merge, doing merge-queue triage, or asking for "quick wins" from the open PR backlog.
bump-dependency
Bumps a Python package dependency across Home Assistant Core integrations, regenerates core requirement files, runs verification tests and prek lint, and prepares a pull request with proper release/compare links.
ha-review
Reviews Home Assistant code changes and provides constructive feedback. Should be used when a review is requested to provide a consistent review behavior and output format. This skill can be used for code reviews in general, not just for GitHub pull requests.
ha-integration-knowledge
Everything you need to know to build, test and review Home Assistant Integrations. If you're looking at an integration, you must use this as your primary reference.
ha-quality-scale-verify
Verifies that a Home Assistant integration follows a specific quality scale rule, checking whether it implements the required patterns, configurations, or code structures defined by the quality scale system. Use when asked to check a rule (e.g. "check if the peblar integration follows the config-flow rule") or to…
ha-pr-comment-audit
Audits the review comment threads on a Home Assistant GitHub pull request, flagging unaddressed comments and requests for clarification. Use when checking whether PR feedback has been handled, either standalone or as part of a full PR review.