SQL Data Analyst
ReviewAudited by ClawScan on May 1, 2026.
Overview
This instruction-only skill matches its SQL analysis purpose, but users should connect only scoped databases because it can run queries, use database credentials, and keep local query/schema records.
Install is reasonable for SQL analysis use. Before using it, connect a limited read-only database account where possible, review SQL before confirming any write/destructive action, and remember that imported data, schemas, saved queries, and query logs may remain in local files.
Findings (3)
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
A generated SELECT query may be run automatically, and write or destructive SQL can be run if the user confirms it.
The configuration allows automatic execution of read-only SELECT queries while requiring confirmation for write operations. This is aligned with the SQL analysis purpose, but it still gives the agent authority to run queries against connected data sources.
"auto_execute_select": true, "require_confirmation_for_writes": true, "show_query_before_execute": true
Review generated SQL before relying on results, keep write confirmations enabled, and use read-only database users for routine analysis.
If broad database credentials are supplied, the agent may be able to read or modify more data than intended.
The skill is designed to use database credentials from environment variables for PostgreSQL and MySQL connections. This is expected for database access, but those credentials determine what data and mutations the agent can perform.
"password_env": "PG_PASSWORD" ... "password_env": "MYSQL_PASSWORD"
Create a least-privilege database account, preferably read-only for analysis, and avoid supplying administrator credentials or broad production access.
Local files may contain database structure or saved analytical context that reveals business data organization.
The skill persists discovered database schema information locally for later reuse. This supports the workflow but may retain sensitive table, column, and relationship information.
Store schema discovery in `./data/schemas/` for reuse.
Periodically review and delete stored schemas, saved queries, imported data, and query logs when they are no longer needed.
