datasource-connectors

Pass

Audited by Gen Agent Trust Hub on May 11, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill consists exclusively of markdown documentation and metadata. It contains no executable scripts, binaries, or automated installation commands.
  • [SAFE]: Explicit security guidelines are integrated into the instructions, specifically mandating the use of parameterized queries to prevent SQL injection and prohibiting the hardcoding of credentials.
  • [SAFE]: Recommends industry-standard practices for handling sensitive data, such as storing API keys in environment variables and utilizing read-only database connections.
  • [PROMPT_INJECTION]: The skill includes instructions to guide the agent's behavior, but these are structured for data format handling and do not attempt to bypass safety filters or override system constraints.
  • [DATA_EXPOSURE]: While the skill defines patterns for REST API authentication and SQL connection strings, it does not contain hardcoded secrets or instructions to exfiltrate data to unauthorized domains.
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to process untrusted external data (CSV, JSON, API responses, etc.), which represents an inherent surface for indirect prompt injection.
  • Ingestion points: External files (CSV, JSON, Excel, Parquet), SQL database results, and REST API responses defined in SKILL.md.
  • Boundary markers: Not specified; instructions focus on parsing logic rather than prompt delimiters.
  • Capability inventory: No code execution or file writing capabilities are included in the skill's own files.
  • Sanitization: Includes requirements for SQL parameterization, but does not specify sanitization for other data formats.
Audit Metadata
Risk Level
SAFE
Analyzed
May 11, 2026, 04:15 PM