Skip to content

charlieviettq/awesome-agent-skill / .claude/skills/algo-hr-compensation

"algo-hr-compensation"

Verified

"Conduct compensation benchmarking analysis to position salaries against market data. Use this skill when the user needs to assess pay competitiveness, build salary bands, or analyze pay equity — even if they say 'are we paying market rate', 'salary benchmarking', or 'compensation analysis'.".

FileKindSize
SKILL.mdskill md4.2 KB
examples/sample_scenario.mdmarkdown5.9 KB
references/band-design.mdmarkdown8.6 KB
references/pay-equity.mdmarkdown13.0 KB

examples/sample_scenario.md

View on GitHub
# Example: SaaS 新創薪酬競爭力稽核

## Scenario

**公司背景:** Telepath AI,台灣 B 輪 SaaS 新創,約 90 名員工,主要產品為企業客服 AI 平台。HR 總監 Jess Chen 正準備 2026 Q2 調薪週期,懷疑工程部門流失率偏高(過去 12 個月離職 6 人)與薪酬低於市場有關。

**可用數據:**
- 內部薪資資料:38 名工程師(含 IC 與 TL),基本薪 NT$1.2M–NT$2.6M
- Mercer Taiwan ICT Survey 2025-H2(2025 年 6 月資料收集)
- 員工性別資料可用,但年資記錄不完整(35% 缺漏)

**用戶問題:**
> 「我想知道我們工程師的薪資競爭力,有沒有人嚴重低於市場?另外請幫我看一下性別薪酬差距。」

---

## Analysis

### Phase 1:Input Validation

**職級對應(Job Matching)**

Telepath 內部職稱對應 Mercer 調查的「軟體開發 / ICT — 台灣中小型企業(< 500 人)」基準:

| 內部職稱 | 人數 | Mercer Benchmark Code | 對應條件 |
|----------|------|-----------------------|----------|
| Junior Engineer | 10 | SW-IC-2 | ≤ 2 年軟體開發經驗,無帶人責任 |
| Engineer | 14 | SW-IC-3 | 3–5 年,獨立交付功能 |
| Senior Engineer | 9 | SW-IC-4 | 6–9 年,跨團隊影響力 |
| Tech Lead | 5 | SW-TL-5 | 帶 3–6 人小組,仍含個人貢獻 |

> **注意:** 1 名頭銜為「Senior Engineer」但實際工作內容偏 Junior 的員工,降階至 SW-IC-2 計算。職稱不等於 job content。

**資料時效確認**

Mercer 2025-H2 資料收集於 2025 年 6 月,距今(2026 年 4 月)10 個月。台灣 ICT 市場 2025 年薪資成長率約 4.2%(Mercer projection)。

老化係數:`1 + 0.042 × (10/12) = 1.035`

**Gate 通過:** 職位已依工作內容對應完畢;調查資料需老化後使用。

---

### Phase 2:Core Algorithm

**市場基準老化後數值(NT$/年,已乘老化係數 1.035)**

| 職級 | P25(老化後) | P50(老化後) | P75(老化後) |
|------|--------------|--------------|--------------|
| SW-IC-2 (Junior) | 1,056,000 | 1,242,000 | 1,449,000 |
| SW-IC-3 (Engineer) | 1,449,000 | 1,759,000 | 2,070,000 |
| SW-IC-4 (Senior) | 1,863,000 | 2,277,000 | 2,691,000 |
| SW-TL-5 (Tech Lead) | 2,277,000 | 2,898,000 | 3,312,000 |

**Compa-Ratio 計算(各職級平均)**

| 職級 | 內部平均實際薪(NT$) | 市場 P50(NT$) | Compa-Ratio |
|------|---------------------|----------------|-------------|
| Junior (n=10) | 1,150,000 | 1,242,000 | **0.93** |
| Engineer (n=14) | 1,560,000 | 1,759,000 | **0.89** |
| Senior (n=9) | 2,050,000 | 2,277,000 | **0.90** |
| Tech Lead (n=5) | 2,780,000 | 2,898,000 | **0.96** |

**個人層級旗標(Below Band < P25)**

- Engineer 職級:3 人基本薪低於市場 P25(NT$1,449,000);其中 1 人低至 NT$1,250,000(compa-ratio 0.71)
- Senior 職級:2 人低於 P25(NT$1,863,000)

合計 **5 人嚴重低於市場(< P25)**,建議列為調薪優先名單。

---

**性別薪酬分析**

| 性別 | 人數 | 平均 Compa-Ratio |
|------|------|-----------------|
| 男性 | 28 | 0.91 |
| 女性 | 10 | 0.88 |

原始差距:女性平均 compa-ratio 低 3.4%。

> **控制合法因素後:** 年資資料缺漏 35%,無法執行完整迴歸控制。以可用的 65% 員工(n=25)進行簡化回歸,控制職級後,性別係數 = −0.028(p = 0.09)。**在目前樣本下未達統計顯著(p < 0.05)**,但缺漏資料本身是風險——需先補齊年資資料才能正式結案。

---

### Phase 3:Verification

- Compa-ratio 分佈:0.71–1.08,整體偏低(正常範圍 0.85–1.15)
- 低於 P25 的 5 人均屬工作年資 3 年以上的 Engineer / Senior,是留才風險最高的族群
- 性別差距現有資料下未顯著,但分析受限於年資缺漏——**結論不完整,需補資料**

**Gate 通過(條件式):** 主要競爭力分析完成;性別薪酬分析需待年資資料補齊後重跑。

---

## Result

```json
{
  "summary": {
    "avg_compa_ratio": 0.91,
    "below_band_pct": 13,
    "above_band_pct": 3
  },
  "by_role": [
    {"role": "Junior Engineer",  "market_p50": 1242000, "avg_actual": 1150000, "compa_ratio": 0.93},
    {"role": "Engineer",         "market_p50": 1759000, "avg_actual": 1560000, "compa_ratio": 0.89},
    {"role": "Senior Engineer",  "market_p50": 2277000, "avg_actual": 2050000, "compa_ratio": 0.90},
    {"role": "Tech Lead",        "market_p50": 2898000, "avg_actual": 2780000, "compa_ratio": 0.96}
  ],
  "priority_adjustments": [
    {"employee_id": "E-017", "role": "Engineer",        "actual": 1250000, "market_p25": 1449000, "gap": -199000},
    {"employee_id": "E-031", "role": "Engineer",        "actual": 1310000, "market_p25": 1449000, "gap": -139000},
    {"employee_id": "E-044", "role": "Senior Engineer", "actual": 1740000, "market_p25": 1863000, "gap": -123000}
  ],
  "equity_flags": [
    {
      "dimension": "gender",
      "gap_pct": 3.4,
      "statistically_significant": false,
      "caveat": "年資資料缺漏 35%,結論不確定;建議補齊後重新分析"
    }
  ],
  "metadata": {
    "employees": 38,
    "survey_source": "Mercer Taiwan ICT Survey",
    "survey_date": "2025-H2",
    "aging_factor": 1.035,
    "analysis_date": "2026-04-10"
  }
}
```

### 行動建議

1. **立即調薪(Q2 週期):** 5 名低於 P25 員工納入優先名單,建議調整至 P40–P50 區間,預估年薪資成本增加約 NT$900K。
2. **Engineer 職級整體偏低(compa-ratio 0.89):** 若預算允許,Engineer band midpoint 建議拉至 NT$1.75M,與市場 P50 對齊。
3. **性別薪酬分析:** 90 天內補齊年資資料,以 Engineer、Senior 兩個職級為優先,重跑迴歸。資料缺漏本身即為 HR 系統的稽核項目。
4. **流失率關聯:** 6 名離職者中,請 HR 確認職級分佈——若離職者集中於 Engineer / Senior,與 compa-ratio 0.89–0.90 高度一致,強化調薪正當性。
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: 472kUpdated Claude CodeCodexUnverifiedInstall
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: 322kUpdated Claude CodeCodexVerifiedInstall
dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependenciesStars: 322kUpdated Claude CodeCodexVerifiedInstall
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 availableStars: 322kUpdated Claude CodeCodexVerifiedInstall