plsql-setup

Pass

Audited by Gen Agent Trust Hub on Aug 26, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION]: The skill facilitates database access, creating an indirect prompt injection surface where untrusted database content could influence agent logic. * Ingestion points: Data retrieved from Oracle database tables via SQLcl or Pythia (SKILL.md). * Boundary markers: The skill mentions that pythia apply provides a journal and preview, but does not define delimiters for LLM context (SKILL.md). * Capability inventory: File system writing and database write access (SKILL.md). * Sanitization: Absent (SKILL.md).
  • [COMMAND_EXECUTION]: The skill references sql -mcp, an Oracle tool for providing an MCP interface to the database (SKILL.md). It recommends the highest restriction level (-R 4) to prevent the agent from executing host-level commands.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 26, 2026, 09:08 AM
Security Audit — agent-trust-hub — plsql-setup