← 返回 Skills 市场

IndexLift SEO Auditor

Name: IndexLift SEO Auditor
Author: horosheff

作者 Artur Horosheff · GitHub ↗ · v1.0.0 · MIT-0

cross-platform ✓ 安全检测通过

394

总下载

当前安装

版本数

在 OpenClaw 中安装

/install google-yandex-seo-skills

功能描述

Use this skill when the user wants an SEO audit, technical SEO review, page-level Google or Yandex analysis, robots.txt or sitemap validation, canonical/inde...

使用说明 (SKILL.md)

IndexLift SEO Auditor

When To Use

Use this skill when the user asks to:

audit a site or landing page for SEO
check technical SEO, on-page SEO, crawlability, or indexability
validate robots.txt, sitemaps, canonicals, redirects, or metadata
analyze a site for Google and Yandex SEO
generate a structured SEO report with priorities and fixes

Do not use this skill for backlink research, SERP monitoring, competitor gap analysis, or any workflow that requires paid external APIs. This build is intentionally limited to the free local tools bundled in the repository.

What This Skill Does

This skill runs an ultra-detailed single-page SEO audit by default and produces:

raw JSON findings
a Markdown report
category scores for technical SEO, on-page SEO, engine-specific signals, and lightweight performance signals
separate Google and Yandex findings
a page snapshot with exact counts for links, headings, images, assets, and metadata
lightweight content-quality heuristics such as opening-copy clarity, paragraph balance, and heading repetition
lightweight local-business heuristics such as contact reachability, messenger presence, conversion paths, and trust markers
a business-oriented priority layer that highlights what to fix first for more leads and stronger trust
a lightweight GEO layer that highlights what already helps AI answer engines, what is missing, and what to fix first
a deeper Yandex layer that explains regional, commercial, legal, markup, and behavioral gaps instead of over-rewarding basic technical presence

Quick Start

From the skill directory, run:

cd .agents/skills/indexlift-seo-auditor
npm install
node scripts/run-audit.js --url "https://example.com" --tier standard --output ./deliverables/

The default mode is single-page. Use --mode crawl only when the user explicitly wants a broader internal-link crawl.

Workflow

Confirm the target URL and desired tier if the user did not specify them.
Run the audit script:

cd .agents/skills/indexlift-seo-auditor
npm install
node scripts/run-audit.js --url "\x3CURL>" --tier standard --engines google,yandex --output ./deliverables/

Read the generated Markdown report and JSON artifact if you need to summarize or further analyze the output.
Present findings ordered by impact:
- direct page failures first
- metadata, heading, and image issues next
- Google/Yandex-specific gaps next
- performance and context-only notes last
Answer in the same language the user is currently using in the conversation unless they explicitly ask for another language.
After giving the short summary, offer the user the full report and mention that the complete Markdown report and JSON artifact are available if they want the full version.

Mode Guidance

single-page: default mode, best for the most detailed audit of one URL
crawl: optional mode for broader internal-link exploration when explicitly needed

Tier Guidance

basic: lightweight single-page audit
standard: full single-page audit with detailed report layers
pro: same single-page default with a broader reporting envelope, but still only free local checks

Output Expectations

When reporting results to the user:

distinguish page-level findings from context-only site signals
distinguish general SEO findings from Google-specific and Yandex-specific findings
do not claim off-page or paid-API data was collected if it was not
treat zero page coverage as a real failure, not a successful audit
prefer concise, actionable fixes over generic SEO advice
call out content clarity, business trust, and conversion friction when those lightweight signals are visible in HTML
distinguish HTML-only GEO heuristics from real-world citation tracking or LLM visibility measurements
keep the final explanation in the same language the user used with the agent unless they explicitly request a different language
after the short answer, offer the full report and tell the user that the complete Markdown report and JSON artifact can be provided or opened on request

Included Files

For install and usage guidance, see references/install.md
For the current check inventory and scoring model, see references/checks.md

安全使用建议

This skill appears to be what it says: a self-contained Node.js SEO auditor that will fetch the target page(s), parse HTML, and write JSON/Markdown reports. Before running it: (1) ensure you run it in a safe environment (it issues HTTP requests to the provided URL — do not point it at internal-only hosts you don't want probed), (2) be aware it performs network requests and will write files to the chosen output directory, (3) confirm your node version meets dependency requirements (some deps may expect a recent Node release) and consider running npm audit on the dependency tree, and (4) if you want to avoid any automated runs, don't grant the agent autonomous execution rights for this skill (the skill itself is not always-enabled, but agent platforms may invoke skills autonomously by default). If you need browser-rendered JS results or Core Web Vitals field data, note the skill explicitly does not provide those.

功能分析

Type: OpenClaw Skill Name: google-yandex-seo-skills Version: 1.0.0 The IndexLift SEO Auditor is a legitimate tool designed for technical and on-page SEO analysis. The codebase, including scripts like 'crawler.js' and 'html-parser.js', focuses on fetching public web content, parsing HTML for SEO signals (metadata, headings, schema.org), and generating reports. It uses standard dependencies (cheerio, fs-extra, fast-xml-parser) and follows the instructions in SKILL.md without any evidence of data exfiltration, credential theft, or malicious prompt injection. While it performs network requests to user-provided URLs, it includes logic to respect robots.txt and limits crawl depth, aligning with its stated purpose.

能力评估

✓ Purpose & Capability

The repo contains a self-contained Node.js SEO auditor (crawler, HTML parser, checks, reporters). The files, dependencies (cheerio, fast-xml-parser, fs-extra), and npm-based usage are coherent with the declared purpose of page-level SEO audits for Google and Yandex. There are no declared env vars or external service credentials that would be out of place for an SEO auditor.

✓ Instruction Scope

SKILL.md instructs the agent to run npm install and node scripts/run-audit.js against a user-provided URL, then read the generated JSON/Markdown artifacts. The instructions do not ask the agent to read unrelated system files, exfiltrate secrets, or call unknown external endpoints. The skill explicitly states it does not use paid APIs and documents what it does and does not measure (for example, it does not perform browser-rendered JS execution). The only external interaction is HTTP(S) fetching of the target site and any public sitemaps/robots links — expected for this purpose.

ℹ Install Mechanism

There is no custom install spec beyond 'npm install' in SKILL.md; package.json and package-lock.json are included. npm will fetch packages from the public registry (common libraries only). This is a typical moderate-risk install pattern (network retrieval from npm) but is proportionate to the Node.js tool's needs. No arbitrary archive downloads, URL shorteners, or personal servers are used in the manifest.

✓ Credentials

The skill does not request environment variables, credentials, or config paths. Its runtime behavior (fetching pages, parsing HTML, saving artifacts) matches that. There are no signs in the code of accessing unrelated secrets or requiring cloud credentials.

✓ Persistence & Privilege

The skill is not always-enabled (always: false) and is user-invocable. It requires node/npm to run but does not attempt to modify other skills or system-wide agent settings. It writes audit artifacts to an output path as part of normal operation.

如何使用

确保已安装 OpenClaw（本地或 Docker 部署）
在对话框中输入安装命令：/install google-yandex-seo-skills
安装完成后，直接呼叫该 Skill 的名称或使用 /google-yandex-seo-skills 触发
根据 Skill 的参数说明提供必要输入，即可获得结构化输出

版本历史

v1.0.0

Initial release of IndexLift SEO Auditor skill. - Provides ultra-detailed single-page SEO audits for Google and Yandex. - Delivers actionable findings with separate Google/Yandex breakdowns in both JSON and Markdown formats. - Includes checks for technical SEO, indexability, metadata, robots.txt, sitemaps, and canonicals. - Offers content-quality and local-business heuristics, as well as business priority and lightweight GEO insights. - Default mode audits one page; internal-link crawling available if explicitly requested. - No external paid APIs required; analysis uses only included local tools.

元数据

Slug google-yandex-seo-skills

版本 1.0.0

许可证 MIT-0

累计安装 0

当前安装数 0

历史版本数 1

常见问题