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 skills add PNGTRID/AnvilWiki --skill anvil-adsense-auditgit clone --depth 1 https://github.com/PNGTRID/AnvilWikiWrote 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/pngtrid/anvilwiki/anvil-adsense-audit)<a href="https://agentmods.dev/skills/pngtrid/anvilwiki/anvil-adsense-audit"><img src="https://agentmods.dev/badge/skills/pngtrid/anvilwiki/anvil-adsense-audit/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/pngtrid/anvilwiki/anvil-adsense-audit"><img src="https://agentmods.dev/badge/skills/pngtrid/anvilwiki/anvil-adsense-audit.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 | $0.00136 | $0.02574 |
| Opus 5.5 | $0.00054 | $0.01030 |
| Sonnet 5.5 | $0.00027 | $0.00515 |
| Haiku 4.5 | $0.00014 | $0.00257 |
Grade A, and why
anvil-adsense-audit scanned grade A with 1 finding 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 1mo 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.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
- 浏览器/curl 实测 sitemap 3 个 URL How it starts
The opening of the file, as written. The whole thing — 102 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AnvilWiki AdSense 申请前体检
回答一个问题:这个站现在交申请,是过、是拒、还是白交? 不凭感觉,逐项体检后给裁决。
核心规则
- Google 官方文档是唯一真相源:AdSense 计划政策、Google 发布商政策、网站质量指南。体检前先刷新一遍官方文档(政策会变);与本文冲突时,官方文档赢。
- 体检不保证过审——它排除的是「明知道会被拒还去交」的浪费;Google 的人工审核永远有主观成分。
- 每项必须四选一:
Pass(有证据)/Fail(有证据)/Unknown(写明缺什么才能判)/N/A(写明为什么不适用)。不许抽样,不许「整体看起来没问题」。 - 具体化:说「正文是抓取摘要且无增量评论」,不说「内容薄」;说「隐私政策没提第三方广告 cookie」,不说「隐私政策不完整」。
模板已解决的项(直接核对,别重复劳动)
这些是模板层承诺,体检时核对存在性即可标 Pass,把它们当免费的:
| 项 | 证据在哪 |
|---|---|
public/ads.txt 就绪授权行存在(注释态——未过审保持注释才是正确状态,占位 pub-ID 反而是无效数据行) |
过审后按文件内注释操作:填入你的发布商 ID、删行首 # 即生效 |
| 隐私政策 / 服务条款 / 版权 / 联系 / 关于 5 个法律页正文内置（非空壳路由），隐私页已含 Google AdSense 广告 cookie 披露段 | src/components/layout/LegalContent.astro；线上 /privacy-policy/ 应有 "Advertising partners (e.g. Google AdSense)" 段 |
| 广告组件 env 门控，默认零广告零弹窗 | PUBLIC_ADSENSE_* 不填就不渲染——申请期天然干净 |
| 空分类列表页 noindex 且不进 sitemap | build 后查 dist/sitemap*.xml |
sitemap 自动生成 + pnpm check-links 死链门禁 |
CI 八门禁 |
| Lighthouse 4×100(速度/体验基线) | 模板开箱契约 |
| 内链尾斜杠 + 语言前缀规范 | pnpm check-content 规则 5/6 |
⚠️ 模板已知缺口(体检时主动提):模板不带 public/robots.txt——Cloudflare Pages 默认对 Googlebot 放行,这本身不是问题;但如果用户自己加过 robots.txt,必须确认没挡 Mediapartners-Google 和 AdsBot-Google,挡了 = 广告抓取失败,典型 Fail 项。
用户侧清单(体检重点,逐项判定)
A. 资格与所有权
- AD-01 自有域名:是自定义域名,还是
*.pages.dev?(后者基本过不了审,见手册课 22「开广告:时机与接入」) - AD-02 所有权信号:站点能验证所有权(GSC / DNS / meta),站点上能看出「谁在运营」(关于页/联系页有真实内容,不是模板占位)。
B. 内容质量(最高频拒因)
- AD-03 篇数:数
src/content/wiki/<主语言>/下非 draft 的.mdx文件,≥15 篇(15-20 篇起申,手册课 22 与 docs/ads.md 口径)。 - AD-04 原创性:抽读 3-5 篇,判断是「自己的话写的攻略」还是「抓取摘要/换词复写/纯视频嵌入无文字」。给出抽读的文件路径和判断依据。
- AD-05 内容深度:抽读页是否覆盖搜索意图(有数据表/步骤/结论),还是 300 字空壳凑数。codes 页有码表即算合格;guides/bosses 页看正文是否答题。
- AD-06 时效性:
codes页lastModified是否在 7 天内(过期码挂「有效」= 内容价值直接判负,跑pnpm refresh-audit看 P0)。 - AD-07 语言一致性:主语言正文是否机翻痕迹明显(不通顺/术语混乱)。模板支持多语言,但申请语言的那个版本必须达到「人类愿意读」的水平。
- AD-08 分类结构:每个分类下都有真实文章(空分类列表页 noindex 了,但导航里挂着 6 个空分类 = 站点未完成信号)。
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.
- 1mo ago First seen · 102 lines · 136 tokens per session scan A fdc98c014651
anvil-adsense-audit is a skill published in the GitHub repository PNGTRID/AnvilWiki (208 stars, last pushed today), licensed MIT. It adds 136 tokens to every session and 2,574 once invoked, about $0.0005 per session on Opus 5.5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-03.
Other skills, from other repositories
data-integrity
What this project is allowed to claim in public — the anchoring rule for AI-generated summaries, confidence states, attribution, and the wording of every published number. Use this skill whenever writing an LLM prompt, rendering generated prose, labelling a metric, designing a badge or status, drafting social posts…
dependency-audit
Audits direct and transitive dependencies for license compliance, maintenance health, CVEs, abandoned packages, and bloat. Triggers on: "audit dependencies", "license check", "dependency health", "abandoned packages", "unused dependencies", "license compliance", "supply chain", "dependency risk".
legal-writer
Legal document drafting -- contracts, memos, briefs, complaints, demand letters, opinions, discovery, settlements, ToS, privacy policies. Full pipeline: document structure, per-section writing, Bluebook citation, case law lookup (CourtListener API), regulation lookup (eCFR API), DOCX output, and TDD-style verification…
contract-playbook
First-pass contract review for {{contractsfolder}}. Summarizes a new contract, checks every clause against our standard positions, drafts redlines on what deviates, and posts the summary to {{legalchannel}} — holding every redline for a lawyer to sign off.
vendor-doc-intake
Daily vendor-onboarding intake for {{requestlabel}}. Validates the W-9, banking form, and signed contract for every new vendor request against a fixed checklist, records every vendor's status in {{vendorregister}}, drafts a follow-up for anything missing or invalid, and posts the batch to {{reviewchannel}}. Load this…
policy-qa
Triage loop for inbound HR policy questions. Classifies each new question from the HR inbox as answerable from the documented Notion policy library or as needing HR's judgment, drafts a grounded Gmail reply for the former, and escalates the latter — anything ambiguous, legal, personal, or compensation-related — to…