classify-domain
GitHub根据域名判断其所属的IAB内容分类，无需获取页面内容。适用于流量补充、黑白名单评分及域名内容识别场景。
触发场景
安装
npx skills add zerogpu/zerogpu-router --skill classify-domain -g -y
SKILL.md
Frontmatter
{
"name": "classify-domain",
"metadata": {
"openclaw": {
"install": [
{
"bins": [
"zerogpu"
],
"kind": "node",
"package": "zerogpu-cli"
}
],
"requires": {
"bins": [
"zerogpu"
]
}
}
},
"description": "Classify a domain name against the IAB taxonomy without fetching the page (zlm-v1-iab-domain-classifier). Use when the user has a hostname or URL rather than article text — bidstream enrichment, allow\/deny-list scoring, \"what is example.com about?\".",
"allowed-tools": "Bash(zerogpu classify_domain*)",
"argument-hint": "<domain>"
}
Sends your input to ZeroGPU's hosted API for inference — this is not local processing. Don't pass secrets, credentials, or regulated data you aren't cleared to share with a third party. See the plugin README's "Data & privacy" section.
Classify a domain:
zerogpu classify_domain "$ARGUMENTS"
Format $ARGUMENTS as a bare hostname. The model takes only the domain — strip the scheme, path, query, and fragment first. https://www.nytimes.com/section/world?x=1 becomes www.nytimes.com. If the user gave you several domains, run the command once per domain.
Output is a JSON object with audience (IAB 2.2 audience tiers with scores) and content (iab_1_0 and iab_2_2 category lists).
Use this instead of the classify-iab skill when the input is a domain — it needs only the hostname, so payloads are up to 10x smaller than sending page text. If the user has the actual article text and wants per-URL precision, use classify-iab or classify-iab-enriched.
Savings note: only if the command output literally contains a line starting with 💰 ZeroGPU savings, append that exact line, unchanged, as the last line of your reply. If no such line is present, say nothing about savings and do not mention or suggest the cost-savings skill — this note is intentionally occasional, not shown every time.
版本历史
- cee9321 当前 2026-08-02 21:15
