gsd-insert-phase

Pass

Audited by Gen Agent Trust Hub on Apr 3, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection as it ingests untrusted data to guide agent actions.
  • Ingestion points: User input provided in {{GSD_ARGS}} (SKILL.md).
  • Boundary markers: No delimiters or isolation markers are used to separate user arguments from instructions.
  • Capability inventory: The workflow invokes spawn_agent for sub-task execution and request_user_input for interaction.
  • Sanitization: Input arguments are used directly without escaping or validation.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 3, 2026, 04:34 AM