tdd-workflows-tdd-cycle

Pass

Audited by Gen Agent Trust Hub on Feb 27, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection via its subagent orchestration logic.
  • Ingestion points: User-provided inputs via the $ARGUMENTS variable are interpolated directly into prompts across all phases of the TDD cycle, including requirements analysis, test generation, and code implementation.
  • Boundary markers: The instructions lack explicit delimiters or guardrail instructions (e.g., "treat the following as data only") to prevent the AI from executing instructions that might be embedded within the $ARGUMENTS data.
  • Capability inventory: The skill utilizes the Task tool to invoke subagents like backend-architect and test-automator, which typically have write access to the filesystem to generate code and tests, creating a path for unauthorized file modification if an injection occurs.
  • Sanitization: There is no evidence of input validation, escaping, or sanitization of the $ARGUMENTS string before it is passed to the sub-orchestration tools.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 27, 2026, 08:56 AM