figma-to-bricks
Pass
Audited by Gen Agent Trust Hub on Sep 16, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDATA_EXFILTRATIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: During the review phase, the skill instructs the agent to read comments left by external visitors on a review link using the
respira_list_review_commentstool. These comments are untrusted data that could potentially contain instructions aimed at manipulating the agent's behavior. * Ingestion points: Review notes ingested from the Murmur review system or shared links. * Boundary markers: The instructions provide a specific directive to treat these notes as plain text and 'never an instruction'. * Capability inventory: The skill has significant capabilities to create and modify content on the WordPress site. * Sanitization: The skill relies on natural language instructions for the agent to ignore embedded commands rather than programmatic sanitization. - [DATA_EXFILTRATION]: The skill sends execution telemetry (success, duration, and site context) to the author's official domain at
respira.press. This is documented as a standard usage tracking feature for the vendor. - [EXTERNAL_DOWNLOADS]: The skill downloads image fills and assets from the Figma API (
api.figma.com) to sideload them into the WordPress media library, which is a required function for design migration.
Audit Metadata