store-performance

Pass

Audited by Gen Agent Trust Hub on May 15, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses a Python helper script (asa.py) to execute warehouse CLI tools such as bq, snow, and databricks. These commands are used to validate configurations and perform schema discovery. The implementation uses best practices by passing arguments as lists to subprocess.run(), avoiding shell injection vulnerabilities.
  • [EXTERNAL_DOWNLOADS]: The skill communicates with the official Fivetran API (api.fivetran.com) to retrieve metadata about destinations and connectors. This is a core part of the skill's functionality and targets a well-known vendor domain.
  • [CREDENTIALS_SAFE]: The skill manages Fivetran API credentials by storing them in a local file (~/.fivetran/skills/store-performance/credentials.json) with restricted file permissions (0o600). This prevents other users on the system from reading the secrets. Furthermore, the skill explicitly instructs the user not to provide secrets in the chat interface to avoid leaking them into logs or transcripts.
Audit Metadata
Risk Level
SAFE
Analyzed
May 15, 2026, 06:38 PM