Figma to Streamlit Component

Pass

Audited by Gen Agent Trust Hub on Apr 9, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection (XSS) through its design-to-code workflow.
  • Ingestion points: Design specifications are ingested from external Figma files using the figma MCP server in SKILL.md.
  • Boundary markers: No explicit delimiters or boundary markers are defined for the fetched design data.
  • Capability inventory: Generated components use st.markdown(..., unsafe_allow_html=True) as seen in ghl_real_estate_ai/streamlit_demo/components/primitives/lead_score_card.py and various templates.
  • Sanitization: The logic lacks sanitization or escaping of the Figma-derived data before it is embedded in HTML strings.
  • [COMMAND_EXECUTION]: The skill documents a "Self-Correction Verification Loop" in SKILL.md that involves the automated execution of generated code within a Playwright environment. This promotes the execution of AI-generated code, which is a potential security risk if the generation instructions are compromised.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 9, 2026, 05:34 PM