api-load-tester

Warn

Audited by Gen Agent Trust Hub on Sep 15, 2026

Risk Level: MEDIUMCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION]: The skill dynamically constructs and executes shell commands using user-provided inputs such as URLs, custom headers, request bodies, and authentication tokens. If these inputs contain shell metacharacters and are not thoroughly sanitized before execution, they could introduce a risk of command injection.
  • [EXTERNAL_DOWNLOADS]: The workflow specifies that if required binaries are missing, the agent should attempt to install them at runtime using brew install hey or go install github.com/rakyll/hey@latest. This introduces unpinned runtime dependencies from external public repositories.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Sep 15, 2026, 01:01 PM
Security Audit — agent-trust-hub — api-load-tester