skills/phuryn/pm-skills/sql-queries/Gen Agent Trust Hub

sql-queries

Pass

Audited by Gen Agent Trust Hub on Sep 15, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze untrusted external data such as database schemas, SQL dumps, and documentation files provided by the user.
  • Ingestion points: The instructions in SKILL.md (Step 1 and Key Capabilities) specify that the agent should read and extract information from uploaded files and diagram descriptions.
  • Boundary markers: There are no instructions or delimiters defined to prevent the agent from following malicious commands that might be embedded as comments or metadata within the uploaded schema files.
  • Capability inventory: The skill generates executable SQL scripts. If the agent is equipped with a database execution tool, an attacker could potentially influence the agent to generate and execute destructive SQL commands (e.g., DROP TABLE) by injecting instructions into the schema files.
  • Sanitization: The skill lacks any explicit validation or sanitization steps for the content of the processed files.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 15, 2026, 12:03 PM
Security Audit — agent-trust-hub — sql-queries