chore: import upstream snapshot with attribution

examples/tools/skills/csv-workbench/SKILL.md

@@ -0,0 +1,20 @@
+---
+name: csv-workbench
+description: Analyze CSV files in /mnt/data and return concise numeric summaries.
+---
+
+# CSV Workbench
+
+Use this skill when the user asks for quick analysis of tabular data.
+
+## Workflow
+
+1. Inspect the CSV schema first (`head`, `python csv.DictReader`, or both).
+2. Compute requested aggregates with a short Python script.
+3. Return concise results with concrete numbers and units when available.
+
+## Constraints
+
+- Prefer Python stdlib for portability.
+- If data is missing or malformed, state assumptions clearly.
+- Keep the final answer short and actionable.

examples/tools/skills/csv-workbench/playbook.md

@@ -0,0 +1,32 @@
+# CSV Playbook
+
+## Quick checks
+
+- Preview rows: `head -n 10 /mnt/data/your-file.csv`.
+- Count rows:
+
+```bash
+python - <<'PY'
+import csv
+
+with open('/mnt/data/your-file.csv', newline='') as f:
+    print(sum(1 for _ in csv.DictReader(f)))
+PY
+```
+
+## Grouped totals template
+
+```bash
+python - <<'PY'
+import csv
+from collections import defaultdict
+
+totals = defaultdict(float)
+with open('/mnt/data/your-file.csv', newline='') as f:
+    for row in csv.DictReader(f):
+        totals[row['region']] += float(row['amount'])
+
+for region in sorted(totals):
+    print(region, round(totals[region], 2))
+PY
+```