lightcraft
Pass
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill invokes local binaries
lightcraftandlightcraft-clito perform photo processing, headless rendering, and application control. - [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from photo libraries and image metadata which could theoretically contain malicious instructions.
- Ingestion points: Reads data through
query_photos,get_develop,ui.inspect, and thelightcraft://resource scheme. - Boundary markers: No explicit delimiters are mentioned for segregating metadata content from instructions.
- Capability inventory: The skill can write files via
app.export,ui.render, andui.screenshot, and executes shell commands via the LightCraft CLI. - Sanitization: There is no explicit mention of sanitizing metadata or library content before processing.
Audit Metadata