open-pencil
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFECOMMAND_EXECUTIONDYNAMIC_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill installs necessary CLI and MCP components (@open-pencil/cli, @open-pencil/mcp) from the author's official package repository for core functionality.
- [DYNAMIC_EXECUTION]: Includes an 'eval' command designed to execute JavaScript code within a Figma Plugin API-compatible runtime, allowing for complex and custom design modifications.
- [COMMAND_EXECUTION]: Operates through a command-line interface ('openpencil') to perform design tasks such as inspecting document structures, querying nodes, and exporting assets.
- [INDIRECT_PROMPT_INJECTION]:
- Ingestion points: The skill processes external design data from '.fig' files and JSX structures provided to the 'render' tool.
- Boundary markers: No specific delimiters or directives are defined to instruct the agent to ignore potentially embedded instructions in the design content.
- Capability inventory: The skill provides capabilities for file system read/write, CLI command execution, and arbitrary script evaluation.
- Sanitization: There is no mention of sanitization or structural validation for the textual content of designs before processing by the agent.
Audit Metadata