db-weapon

Pass

Audited by Gen Agent Trust Hub on Jul 25, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill includes shell and SQL scripts designed to perform database audits. These scripts, such as analyze-query-plan.sh, audit-missing-indexes.sql, and bloat-check.sql, are intended to be executed against a PostgreSQL environment to retrieve performance metrics, identify missing indexes, and check for table bloat.
  • [EXTERNAL_DOWNLOADS]: The documentation and research notes cite numerous external technical resources. These include official documentation from postgresql.org, pgbouncer.org, and timescale.com, as well as repositories and websites for well-known database platforms and tools like Supabase, Neon, Drizzle, and Prisma. These links are provided for informational and citation purposes.
  • [PROMPT_INJECTION]: The skill is designed to analyze existing database schemas and project files, which creates an attack surface for indirect prompt injection.
  • Ingestion points: The agent is instructed to read local files such as schema.sql, package.json, and ORM schema files (schema.ts, schema.prisma) to perform reviews.
  • Boundary markers: The instructions do not specify the use of specific delimiters or "ignore embedded instructions" warnings when reading these files.
  • Capability inventory: The skill has the capability to execute shell scripts (analyze-query-plan.sh) and SQL commands via psql based on its analysis.
  • Sanitization: There are no explicit sanitization or filtering steps defined for the content read from the database schema or configuration files.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 25, 2026, 03:22 AM
Security Audit — agent-trust-hub — db-weapon