atomic-page-builder
Pass
Audited by Gen Agent Trust Hub on Mar 1, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection as it requires the agent to ingest and act upon data from an external, potentially untrusted source (Figma).\n
- Ingestion points: The agent is instructed to fetch page specifications, layout values, and CSS snippets from Figma nodes using the
figma.get_nodetool and Dev Mode features, as documented inSKILL.extended.mdandreferences/figma-extraction-guide.md.\n - Boundary markers: The skill lacks requirements for boundary markers or delimiters to differentiate between layout data and potential embedded instructions within the Figma data.\n
- Capability inventory: The agent has the capability to generate new React and CSS code, as well as modify existing project configuration files such as
atomic-design/tokens/*.json.\n - Sanitization: There is no mention of sanitizing or validating the data retrieved from Figma before it is integrated into the codebase or used to overwrite existing design tokens.
Audit Metadata