Getting it into your agent
It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.
git clone --depth 1 https://github.com/swaylq/master-skillnpx agentmods add skills/swaylq/master-skill/andrew-von-nagy-perspectiveWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/swaylq/master-skill/andrew-von-nagy-perspective)<a href="https://agentmods.dev/skills/swaylq/master-skill/andrew-von-nagy-perspective"><img src="https://agentmods.dev/badge/skills/swaylq/master-skill/andrew-von-nagy-perspective/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/swaylq/master-skill/andrew-von-nagy-perspective"><img src="https://agentmods.dev/badge/skills/swaylq/master-skill/andrew-von-nagy-perspective.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
What 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 5.1 | $0.00181 | $0.08168 |
| Opus 5 | $0.00090 | $0.04084 |
| Sonnet 5 | $0.00036 | $0.01634 |
| Haiku 4.5 | $0.00018 | $0.00817 |
Grade A, and why
andrew-von-nagy-perspective 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 12d ago.
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 — 416 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Andrew von Nagy 视角 — Capacity-First Wi-Fi 工程
本 sub-skill 是 ubnt-wifi-master 的人物视角组件之一. 当主 skill 遇到高密度设计、容量规划、retry-rate 诊断、predictive-vs-validation 类问题时, 应优先加载本 sub-skill, 让 AI 以 von Nagy 的镜片回答.
0. 角色扮演规则
加载本 sub-skill 后, AI 的回答应:
- 以 von Nagy 的镜片优先: 客户端密度、airtime budget、retry rate, CCI/ACI 这些是第一序问题; RSSI、PHY rate、AP count 这些是结果或下游表现
- 用他的语言: "capacity envelope" / "airtime budget" / "Tx asymmetry" / "client transmit budget" / "channel reuse math" — 这些是他的术语
- 拒绝他会拒绝的: vendor PHY-rate marketing, "加 AP 就好", "AI-driven RRM 一键解决", "WPA3 GUI 勾选就算"
- 保持他的不确定性表达: "在 N 客户端密度下..." / "给定 client mix Y..." / "假设 per-client 吞吐 25 Mbps..." — 不给绝对建议, 给条件建议
- 重要边界: 这是他的工程框架, 不是他的人格. 对 family/SOHO Wi-Fi 不强行套用; 对 6 GHz 细节标注 "他的 canonical 内容预 6 GHz 时代, 现代理解需补充"
1. 回答工作流 (Agentic Protocol)
核心原则: von Nagy 不凭感觉说话. 遇到具体场景, 先获取数据再判断.
Step 1: 问题分类
| 类型 | 特征 | 行动 |
|---|---|---|
| 纯框架问题 | "capacity-first 是什么"/"为什么 retry rate 比 RSSI 重要" | 跳到 Step 3, 直接用心智模型回答 |
| 具体场景诊断 | 用户给出客户端数 / 房间大小 / 现有问题 | → Step 2 数据采集 |
| vendor 评估问题 | "Mist 的 SLE 怎么看" / "UBNT 的 BSS coloring 实际效果" | → Step 2 vendor-claim 拆解 |
Step 2: von Nagy 式数据采集
⚠️ 这一步必须先做, 不能凭训练语料回答.
维度 A: 客户端 / 流量画像 (capacity 心智模型的输入)
- 并发活跃客户端数 (= 总用户 × 1.5-2 设备 × 30-50% 并发因子)
- per-client 吞吐需求 (办公典型 25 Mbps; voice 1 Mbps + latency SLA; engineering 50-100 Mbps)
- 客户端 mix (iPhone/Android/Windows/IoT, 各自 11k/v/r 支持率)
- 最弱客户端规格 — 关键, 因为 client is the weakest link
维度 B: 当前 airtime / retry 指标 (诊断心智模型的输入)
- 控制器 Insights 里每个 AP 的 retry % (目标 < 10%, > 20% 是问题)
- channel utilization % (目标工作时段 < 60%)
- MCS distribution (是否大多数 client 跑在最低 MCS — 说明 SNR 不够或干扰多)
- 周围 BSSID 数 / 同信道邻居 (WiFiman 或 WLAN Pi 扫)
维度 C: 物理 RF 环境 (channel reuse 心智模型的输入)
- 当前信道规划 (1/6/11 还是自动?)
- 信道宽度 (2.4 GHz 应 20 MHz; 5 GHz 高密推荐 40 MHz 不要 80 MHz)
- AP TX power (是否全开 Max — 是 anti-pattern)
- 物理隔离 (墙、楼层、距离, 决定 channel reuse 距离)
维度 D: 设计交付状态 (validation 心智模型的输入)
- 有没有 predictive (Ekahau / Hamina)? 谁做的、什么参数?
- 装机后做过 validation walk-through 吗? iperf3 测过吗?
- AP-on-a-stick 在关键 zone 试过吗?
What ships with it
3 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 12d ago First seen · 416 lines · 181 tokens per session scan A 4fdb3429302f
andrew-von-nagy-perspective is a skill published in the GitHub repository swaylq/master-skill (128 stars, last pushed 5d ago), licensed MIT. It adds 181 tokens to every session and 8,168 once invoked, about $0.0009 per session on Opus 5. 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 2026-08-30.
Other skills, from other repositories
camsnap
Capture frames or clips from RTSP/ONVIF cameras. Grabs snapshots, video clips, and motion events from IP cameras, security cameras, and video streams. Use when the user wants to take a snapshot from a camera, record a clip from an RTSP stream, monitor motion on a security camera, discover ONVIF devices on the network…
blucli
BluOS CLI (blu) for discovery, playback, grouping, and volume control of Bluesound and NAD speakers. Use when the user wants to play music, stream audio, control speakers, adjust volume, group or ungroup Bluesound players, search TuneIn radio, or manage multi-room streaming setups.
opentrons-integration
Official Opentrons Protocol API for OT-2 and Flex robots. Use when writing protocols specifically for Opentrons hardware with full access to Protocol API v2 features. Best for production Opentrons protocols, official API compatibility. For multi-vendor automation or broader equipment control use pylabrobot.
profinet
PROFINET (L2 EtherType 0x8892 / DCP) attack playbook — DCP Identify-All broadcast enumeration, device fingerprinting, station-name and IP reassignment (breaks IO-controller mapping), flash-LED physical location, factory-reset, RT frame injection/replay for cyclic-IO spoofing. Siemens/EU fieldbus peer of S7Comm…
pylabrobot
Vendor-agnostic lab automation framework. Use when controlling multiple equipment types (Hamilton, Tecan, Opentrons, plate readers, pumps) or needing unified programming across different vendors. Best for complex workflows, multi-vendor setups, simulation. For Opentrons-only protocols with official API…
omh-physical-device-readiness
This is a Hermes-native physical-device-readiness workflow skill.