skills/sendaifun/skills/sanctum/Gen Agent Trust Hub

sanctum

Pass

Audited by Gen Agent Trust Hub on Feb 16, 2026

Risk Level: LOWPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [Indirect Prompt Injection] (LOW): The skill documentation includes examples for fetching token metadata from an external API (sanctum-api.ironforge.network). An attacker who compromises this API or its data could inject instructions to influence the agent. * Ingestion points: resources/lst-reference.md via API fetch examples. * Boundary markers: None provided in the documentation. * Capability inventory: The snippets describe network requests (fetch) and account data retrieval. * Sanitization: None described in the code snippets.
  • [Unverifiable Dependencies & Remote Code Execution] (LOW): The documentation references the igneous-labs GitHub organization and provides instructions for using solana-verify with these repositories. This organization is not among the pre-approved trusted sources.
Audit Metadata
Risk Level
LOW
Analyzed
Feb 16, 2026, 02:47 AM