NYC
skills/smithery/ai/test-suite-create/Gen Agent Trust Hub

test-suite-create

Fail

Audited by Gen Agent Trust Hub on Feb 16, 2026

Risk Level: HIGHPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [Indirect Prompt Injection] (HIGH): The skill possesses a high-risk attack surface for indirect prompt injection.
  • Ingestion points: Processes 'target component or service' (source code) in Step 1.
  • Boundary markers: Absent; no instructions to ignore embedded instructions in the source code.
  • Capability inventory: Step 3/4 involves file-writing (*.test.ts), and Step 5 involves arbitrary command execution via a test runner (Jest/Vitest).
  • Sanitization: Absent; the skill does not mention validating or escaping content from the target component before implementing tests.
  • [Command Execution] (MEDIUM): The skill is designed to execute external commands ('Run tests') using system tools. While standard for a developer tool, this capability can be abused if the input content is manipulated to include shell metacharacters or malicious test logic.
Recommendations
  • AI detected serious security threats
Audit Metadata
Risk Level
HIGH
Analyzed
Feb 16, 2026, 09:55 AM