lightcraft-editor
Pass
Audited by Gen Agent Trust Hub on Nov 4, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to execute a local Node.js adapter script (
node ui/dist/adapter.mjs --serve) via the terminal if the editor service is not already registered. This is a functional requirement for setting up the local development environment for the LightCraft editor. - [INDIRECT_PROMPT_INJECTION]: The skill facilitates interaction with external workspace data and editor outputs, creating a potential surface for indirect prompt injection.
- Ingestion points: Workspace paths provided to
open_editorand files handled viaopen_file. - Boundary markers: The instructions do not specify any delimiters or safety warnings for processing embedded content.
- Capability inventory: Includes terminal command execution (
node) and tool calls for inspecting and running commands (inspect,command_run). - Sanitization: No specific sanitization or content validation steps are documented in the skill instructions.
Audit Metadata