swap-instance-overrides
Pass
Audited by Gen Agent Trust Hub on Sep 10, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructions involve reading text content from Figma layers using
scan_text_nodesandget_node. This creates a surface where the agent could ingest data containing malicious instructions from the design environment.\n - Ingestion points: Data is ingested through
get_selection(),get_node(), andscan_text_nodes()inSKILL.md.\n - Boundary markers: The instructions do not specify any delimiters or safety warnings for the agent to ignore instructions embedded in the design text.\n
- Capability inventory: The skill uses tools for Figma manipulation (
set_text,set_fills,set_strokes) but does not include system-level execution or network exfiltration tools.\n - Sanitization: There are no mentions of sanitizing or validating the text content read from the source instances before processing it.
Audit Metadata