diff --git a/skills/quantitative-resume-auditor/SKILL.md b/skills/quantitative-resume-auditor/SKILL.md
new file mode 100644
index 00000000..1985cce3
--- /dev/null
+++ b/skills/quantitative-resume-auditor/SKILL.md
@@ -0,0 +1,88 @@
+<!--
+SPDX-License-Identifier: Apache-2.0
+Copyright 2026 Dr. Adriano De Souza
+-->
+
+---
+name: quantitative-resume-auditor
+description: Audits any resume using Python to compute metric density (M/B ratio). Benchmarks against a 3.0 Gold Standard. Triggers on resume audit, density check, impact scoring, or JD gap analysis.
+---
+
+# Quantitative Resume Auditor
+
+This skill transforms qualitative resume reviews into high-precision computational audits producing a fixed Metric Density Score calculated by Python.
+
+**Core formula:** Density = Total Numeric Metrics (M) / Total Bullet Points (B)
+
+A score of 3.0+ means the resume is data-dense. Works on any resume — any length, profession, or career stage.
+
+See REFERENCE.md for benchmarks and JD coverage framework.
+
+---
+
+## Instructions for Claude
+
+### Step 1 — Data Extraction
+
+Count three variables precisely from the resume:
+
+| Variable | Definition |
+|----------|------------|
+| B | Total bullet points |
+| M | Total numeric metrics (numbers, %, $, ratios) |
+| Q | Bullets with at least one metric |
+
+Count M by category: Percentages, Currency, Headcount, Years, Score gains, Efficiency, Volume.
+
+### Step 2 — Python Computation
+
+Run using Claude's code execution tool:
+
+```python
+def run_resume_audit(M, B, Q):
+    if B == 0:
+        return {"Error": "No bullet points detected."}
+    density = round(M / B, 2)
+    rate = min(round((Q / B) * 100, 1), 100.0)
+    if density >= 3.0:
+        status, rec = "GOLD STANDARD", "Maintain density. Focus on JD gaps."
+    elif density >= 1.5:
+        status, rec = "COMPETITIVE", "Add 1-2 metrics to the 5 weakest bullets."
+    else:
+        status, rec = "QUALITATIVE", "Rewrite responsibilities as accomplishments."
+    return {"Total Metrics (M)": M, "Bullet Points (B)": B,
+            "Quantified Bullets (Q)": Q, "Density": f"{density}/bullet",
+            "Rate": f"{rate}%", "Status": status, "Recommendation": rec}
+
+M, B, Q = 0, 0, 0  # Claude replaces with actual counts
+for k, v in run_resume_audit(M, B, Q).items():
+    print(f"{k}: {v}")
+```
+
+### Step 3 — Report Format
+
+1. **Audit Scoreboard** — table with M, B, Q, Density, Rate, Status
+2. **Category Breakdown** — per-category metric counts
+3. **Gap Analysis** — JD coverage map (if JD provided) or weakest bullet list
+
+---
+
+## When to Use
+
+- Resume audit, score, density check, or impact analysis
+- "Is my resume quantitative enough?"
+- JD gap analysis or coverage map
+- Career coaching on resume strength
+
+## Example Prompts
+
+- "Audit this resume for data density."
+- "Score my resume against the 3.0 benchmark."
+- "Run a gap analysis — here is my resume and the job description."
+
+## Notes
+
+- **Dynamic:** No fixed target — counts whatever is in the document.
+- **Requires code execution** enabled in Claude.
+- **3.0 benchmark** is a professional heuristic. See REFERENCE.md.
+- **Any profession:** education, business, healthcare, tech, government, nonprofit.