create-skill-glyph
Pass
Audited by Gen Agent Trust Hub on Mar 18, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses Bash to execute Rscript commands for rendering icons, which is a necessary part of the R-based visualization workflow.
- [EXTERNAL_DOWNLOADS]: The instructions suggest installing standard R packages (ggplot2, ggforce, ggfx, ragg, magick) from official CRAN mirrors, which are trusted sources for R development.
- [PROMPT_INJECTION]: The skill ingests data from external SKILL.md files to generate R code, creating a surface for indirect prompt injection. (1) Ingestion points: Reads SKILL.md files to understand visual concepts in Step 1. (2) Boundary markers: No explicit delimiters or instructions to ignore embedded commands are defined for the input text. (3) Capability inventory: The agent is tasked with writing R code and executing it via Bash/Rscript in Steps 2 and 5. (4) Sanitization: No explicit sanitization or validation of the ingested concept text is described before its use in code generation.
Audit Metadata