shopify-app-release-assistant

Pass

Audited by Gen Agent Trust Hub on Feb 19, 2026

Risk Level: SAFENO_CODEPROMPT_INJECTION
Full Analysis
  • [NO_CODE] (SAFE): The skill consists entirely of Markdown documents (SKILL.md and reference files). No scripts, executables, or package manager configurations are present.
  • [PROMPT_INJECTION] (SAFE): An indirect prompt injection surface exists as the skill processes user-supplied app descriptions and test credentials. This is rated SAFE because the skill possesses no tools or capabilities to execute commands or exfiltrate data. 1. Ingestion points: User app descriptions and test credentials in SKILL.md. 2. Boundary markers: Absent. 3. Capability inventory: None; limited to text generation. 4. Sanitization: Absent.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 19, 2026, 03:37 AM