skills/kmalakoff/sensemaking/mdsql/Gen Agent Trust Hub

mdsql

Pass

Audited by Gen Agent Trust Hub on Aug 8, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill suggests installing the mdsql package from NPM if the tool is not found on the system. This package is an official resource from the skill author.
  • [COMMAND_EXECUTION]: The agent is instructed to use the mdsql CLI to perform various tasks such as listing queries, running SQL queries, and managing the local cache.
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests data from markdown frontmatter in the user's workspace. While this is a potential attack surface for indirect prompt injection, the skill explicitly mentions using SQL placeholders (?) for named queries to bind parameters securely, which follows security best practices.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 8, 2026, 09:25 PM
Security Audit — agent-trust-hub — mdsql