mysql-tools

Warn

Audited by Gen Agent Trust Hub on Mar 6, 2026

Risk Level: MEDIUMCOMMAND_EXECUTIONCREDENTIALS_UNSAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION]: The script scripts/mysql_query.py executes arbitrary SQL strings provided via command-line arguments, giving the agent full control over the database including data modification and deletion.
  • [COMMAND_EXECUTION]: The script scripts/mysql_schema.py is vulnerable to SQL injection because it uses string interpolation to insert the table name directly into a query (DESCRIBE {table}).
  • [CREDENTIALS_UNSAFE]: All scripts require database passwords to be passed as command-line arguments, which can expose them in system process lists and shell history.
  • [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection. 1. Ingestion points: External data entering the agent's context. 2. Boundary markers: None present. 3. Capability inventory: Full database access via mysql_query.py. 4. Sanitization: No input validation or escaping on queries.
  • [EXTERNAL_DOWNLOADS]: The skill requires the pymysql package from PyPI, which is a well-known and trusted package registry.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Mar 6, 2026, 04:13 PM