skills/horizon-foundry/foundry/mobile/Gen Agent Trust Hub

mobile

Pass

Audited by Gen Agent Trust Hub on Oct 8, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill includes instructions to perform a version check by fetching metadata from https://foundry.thehorizonfoundry.com/api/version. This request targets a vendor-owned domain and is used to notify the user of available updates.
  • [COMMAND_EXECUTION]: The skill provides a shell script for the agent to manage version checks. This script uses find to check for a rate-limit stamp file in the user's home directory and curl to query the vendor's API. The command is designed for information retrieval and does not execute remote content.
  • [INDIRECT_PROMPT_INJECTION]: The diagnostic process involves inspecting untrusted content from a mobile web surface, such as element focus states and computed styles. This data is ingested into the agent's context to guide UI fixes.
  • Ingestion points: DOM properties and methods (e.g., document.activeElement, getComputedStyle) in SKILL.md.
  • Boundary markers: None identified.
  • Capability inventory: A shell command using curl for version checking is present in SKILL.md.
  • Sanitization: No specific sanitization or escaping of DOM data is mentioned before processing.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 8, 2026, 03:02 PM