load-testing

Pass

Audited by Gen Agent Trust Hub on May 16, 2026

Risk Level: SAFECREDENTIALS_UNSAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [CREDENTIALS_UNSAFE]: The skill instructs the agent to solicit sensitive credentials, specifically Databricks M2M OAuth client IDs and secrets, using the AskUserQuestion tool. While these are required for authenticating load tests against the user's workspace, capturing them via prompt exposure introduces them into the agent's conversation history.
  • [COMMAND_EXECUTION]: The skill facilitates the execution of various CLI tools and orchestration scripts. This includes managing Databricks App deployments via databricks bundle, managing Python environments with uv, and running the Locust load-testing framework and dashboard generation scripts.
  • [EXTERNAL_DOWNLOADS]: The instructions require the installation of third-party Python packages (locust, requests, urllib3) from the public PyPI registry. These are well-known libraries standard for performance testing and HTTP operations.
Audit Metadata
Risk Level
SAFE
Analyzed
May 16, 2026, 06:08 AM
Security Audit — agent-trust-hub — load-testing