Skills
Navigate
skills/gabelul/stitch-kit/stitch-mcp-generate-variants/Gen Agent Trust Hub

stitch-mcp-generate-variants

Pass

Audited by Gen Agent Trust Hub on Sep 20, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill uses a prompt argument in the generate_variants tool to influence design output based on user instructions.
  • Ingestion points: The prompt argument in the generate_variants tool call (SKILL.md).
  • Boundary markers: Absent; user-provided text is passed directly as a string argument to the tool without explicit delimiters or instructions to ignore embedded commands.
  • Capability inventory: The skill is authorized to use the stitch*:* toolset, allowing it to list, retrieve, and modify design assets (SKILL.md).
  • Sanitization: No specific sanitization or validation routines for the prompt input are described within the skill's instructions.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 20, 2026, 09:01 PM