charlieviettq/awesome-agent-skill / .claude/skills/algo-seo-backlink
"algo-seo-backlink"
Verified"Evaluate backlink quality using Domain Authority, Domain Rating, and trust metrics. Use this skill when the user needs to assess link profile health, identify toxic backlinks, or plan link building strategy — even if they say 'check my backlinks', 'link building', or 'domain authority analysis'.".
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 4.1 KB |
| examples/sample_scenario.md | markdown | 6.0 KB |
| references/competitor-gap.md | markdown | 8.3 KB |
| references/toxicity-scoring.md | markdown | 9.1 KB |
examples/sample_scenario.md
View on GitHub# Example: 台灣電商網站反向連結健檢
## Scenario
**用戶問題:**
> 我們是台灣中型家具電商「木質家居」(mokuzai.tw),近三個月自然搜尋流量下降 22%。SEO 顧問懷疑是 Google 懲罰導致,請幫我們評估反向連結品質,找出需要 disavow 的有毒連結,並評估整體連結輪廓健康狀況。
**背景資訊:**
- 網站上線 4 年,主要商品:實木餐桌、沙發、收納家具
- 2024 年底曾委託外包公司做「SEO 衝排名」,事後發現對方使用批量建連結
- 從 Ahrefs 匯出 487 條反向連結,涵蓋 234 個 referring domains
- 匯出日期:2025-03-10
---
## Analysis
### Phase 1:Input Validation
匯出資料確認欄位完整:
| 欄位 | 狀態 |
|------|------|
| Referring domain | ✅ |
| Domain Rating (DR) | ✅ |
| Anchor text | ✅ |
| Link type (do/nofollow) | ✅ |
| First seen date | ✅ |
**Gate passed** — 487 條連結,234 個 referring domains,資料可進行分析。
---
### Phase 2:Core Algorithm
**步驟 1:依 referring domain 去重**
487 條原始連結 → 234 個唯一 referring domains(部分站點貢獻多條連結,去重後以最高 DR 連結代表)。
**步驟 2:DR 分佈評分**
| DR 區間 | Domains 數 | 佔比 | 評分 |
|---------|-----------|------|------|
| DR 60+ | 18 | 7.7% | 高品質 |
| DR 20–59 | 67 | 28.6% | 中等 |
| DR 1–19 | 89 | 38.0% | 低品質 |
| DR 0 (新站/死站) | 60 | 25.6% | 可疑 |
平均 DR:**21.3**(台灣電商平均水準約 28–35,偏低)
**步驟 3:有毒連結標記**
篩選條件:
1. DR < 10 且連結超過 3 條(批量建連結特徵)
2. 非中文、非英文語系(越南語、俄語垃圾連結農場)
3. 網域含 `bestlinks`、`seolinkfarm`、`guestpost99` 等已知 PBN 名稱模式
4. 連結頁面標題與家具完全無關(如:博弈、藥品、成人內容)
標記結果:**41 個 referring domains 列為有毒**
| 類型 | 數量 | 典型特徵 |
|------|------|---------|
| 連結農場 (link farm) | 19 | DR 0–3,頁面只有連結清單,無實質內容 |
| PBN 網域 | 12 | DR 8–15,多個網域指向同一 IP 段 |
| 無關語系垃圾站 | 7 | .ru / .vn 網域,頁面為俄文/越文 |
| 已下線但仍被計算的死站 | 3 | HTTP 500,但 Ahrefs 仍顯示歷史連結 |
**步驟 4:錨文字分佈分析**
| 錨文字類型 | 佔比 | 健康基準 | 狀態 |
|-----------|------|---------|------|
| 品牌名(木質家居、mokuzai) | 12% | 30–50% | ⚠️ 偏低 |
| 裸網址(mokuzai.tw) | 8% | 15–25% | ⚠️ 偏低 |
| 精確關鍵字(實木餐桌推薦) | 41% | < 10% | 🔴 嚴重過高 |
| 部分關鍵字(餐桌、家具) | 28% | 10–20% | ⚠️ 偏高 |
| 雜項(點此、更多) | 11% | 10–20% | ✅ 正常 |
**診斷**:精確關鍵字錨文字達 41%,遠超安全值 (<10%),這是人工建連結的典型特徵,與外包 SEO 公司的操作高度吻合。這是觸發 Google 懲罰的主要嫌疑。
---
### Phase 3:Verification
**交叉驗證**:
- 將 41 個有毒 domains 比對 Moz Spam Score > 60 的黑名單,確認 37 個重疊
- 4 個 DR 低但未出現在黑名單的 domains 進行人工覆核:
- 2 個確認為新創小型部落格,非垃圾站 → 移出 disavow 清單
- 2 個確認為連結農場 → 維持 disavow
**最終有毒清單:39 個 referring domains**
**Gate passed** — Anchor profile 異常確認,有毒連結清單核實完成。
---
### Phase 4:連結輪廓健康總結
**時間序列異常**:
檢查 first seen date — 2024 年 10 月至 11 月間出現 **單月新增 78 個 referring domains**,遠超正常月均 5–8 個。這個爆量時間點與外包 SEO 操作時間吻合,也是 Google 演算法更新觸發懲罰的典型信號。
---
## Result
```json
{
"profile": {
"referring_domains": 234,
"avg_dr": 21.3,
"toxic_count": 39,
"anchor_diversity": 0.31,
"dr_distribution": {
"dr_60_plus": "7.7%",
"dr_20_59": "28.6%",
"dr_1_19": "38.0%",
"dr_0": "25.6%"
},
"exact_match_anchor_ratio": "41%",
"health_score": "Poor — Penalty Risk High"
},
"actions": [
{
"type": "disavow",
"domains": [
"bestlinks99.ru",
"seofarm-tw.com",
"guestpost-asia.vn",
"... (共 39 個 domains)"
],
"reason": "link farm pattern / PBN / 無關語系垃圾站",
"priority": "high",
"deadline": "2025-03-20 前提交 Google Disavow Tool"
},
{
"type": "anchor_text_rebalancing",
"target": "精確關鍵字錨文字降至 10% 以下",
"method": "停止任何人工建連結;透過 PR 與媒體曝光獲取品牌錨文字自然連結",
"priority": "high"
},
{
"type": "link_building",
"focus": "DR 40+ 台灣家居 / 設計 / 生活風格媒體",
"targets": ["La Vie 行動家", "Shopping Design", "品味台灣生活"],
"method": "內容合作、產品評測、專家專欄",
"priority": "medium"
}
],
"risk_flags": [
"2024-10 單月爆量 78 個 referring domains — Google spam filter 觸發風險",
"41% exact-match anchor ratio — 遠超安全值,演算法懲罰主因",
"DR 0 死站佔 25.6% — 拉低整體輪廓品質"
],
"metadata": {
"tool": "ahrefs",
"export_date": "2025-03-10",
"analysis_date": "2025-03-12",
"analyst": "algo-seo-backlink"
}
}
```
**行動摘要:**
1. **立即(本週)**:提交 39 個 domains 的 disavow 檔案至 Google Search Console
2. **短期(1–3 個月)**:停止所有人工建連結,等待 Google 重新爬取並生效
3. **中期(3–6 個月)**:透過真實內容合作取得 5–10 條 DR 40+ 高品質連結,稀釋現有不健康錨文字比例
4. **預期**:Disavow 生效約需 2–3 個月;流量恢復需 4–6 個月,視 Google 重新評估時間而定
> **注意**:DA/DR 是 Ahrefs/Moz 的第三方估算值,非 Google 官方指標。Disavow 檔案提交後無法即時看到效果,切勿因等待焦慮而重複提交或額外操作。
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis. | Stars: 472k | Updated | Claude CodeCodex | Unverified | Install |
| diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness. | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is available | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |