alloy-howtos

Pass

Audited by Gen Agent Trust Hub on Apr 13, 2026

Risk Level: SAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The documentation references downloading development tools from official registries like npmjs.org and the tidev organization on GitHub. These are legitimate, well-known sources for the Titanium SDK community.
  • [COMMAND_EXECUTION]: The skill is configured to allow 'alloy' and 'node' commands via the Bash tool. This capability is strictly scoped and necessary for the skill's purpose of creating, compiling, and managing Titanium Alloy projects.
  • [PROMPT_INJECTION]: The skill has a surface for indirect prompt injection as it processes project files that could contain embedded instructions. \n- Ingestion points: Project configuration and source files (e.g., 'app/' folder, 'alloy.jmk', 'config.json'). \n- Boundary markers: None are defined to separate ingested file content from the agent's instructions. \n- Capability inventory: Can execute 'alloy' and 'node' commands, and has permissions to edit or write files. \n- Sanitization: No specific sanitization of ingested project content is mentioned before processing.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 13, 2026, 12:26 AM