explain-with-anim
Pass
Audited by Gen Agent Trust Hub on Oct 2, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill interacts with untrusted data by analyzing local repository content (source code, directory structures, and Markdown files) to generate visualizations, which presents a surface for indirect prompt injection.
- Ingestion points: The tool ingests data via
vlmkit-anim facts(directory analysis),vlmkit-anim repo(workspace analysis), andvlmkit-anim import mermaid(extracting diagrams from Markdown files). - Boundary markers: The instructions do not define specific delimiters or instructions for the agent to ignore embedded commands or prompt-like text within the analyzed repository files.
- Capability inventory: The skill uses the
vlmkit-animtool to read the filesystem and write output files including JSON scene definitions, SVGs, and GIFs (SKILL.md). - Sanitization: The skill incorporates a validation step using
vlmkit-anim checkto ensure the generated diagrams match expected code facts, though it does not explicitly define sanitization for the raw input data itself.
Audit Metadata