Setup & Installation
Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:
What This Skill Does
Parses employee skills CSV files and generates three idempotent SQL scripts to populate skill_categories, skills, and employee_skills tables in a Supabase database. Handles name fuzzy matching, deduplication, and whitespace normalization automatically.
Generating idempotent scripts with fuzzy name correction and deduplication eliminates the manual work of reconciling CSV names against database records and prevents common constraint violations.
When to use it
- Importing bulk employee skill data from an HR spreadsheet export
- Updating years_of_experience records without overwriting unrelated rows
- Syncing a new CSV after onboarding without re-running the full migration
- Catching name spelling mismatches between CSV exports and the database
- Generating auditable SQL scripts for DBA review before execution
Example Workflow
Here's how your AI assistant might use this skill in practice.
User uploads employee_skills.csv and says: Parse this file and generate SQL insert scripts
- 1Parse row 2 for skill category names and subsequent rows for individual skill names, mapping each to its column position
- 2Query Supabase skill_categories, skills, and employees tables to determine what already exists
- 3Generate 1_insert_categories.sql with ON CONFLICT DO NOTHING for missing categories
- 4Generate 2_insert_skills.sql with subquery-based category_id lookups for missing skills
- 5Fuzzy-match CSV employee names against the database, deduplicate employee-skill pairs keeping the highest experience value, generate 3_insert_employee_skills.sql and skipped_employees_report.txt
Three SQL files ready to execute in order (1 → 2 → 3) plus a report listing auto-corrected names and skipped employees
Requirements
Accounts, API keys, or tools you or your AI assistant may need to set up while using this skill.