manim-video
Pass
Audited by Gen Agent Trust Hub on Sep 5, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
- [DYNAMIC_EXECUTION]: The core functionality of the skill involves generating Python scripts (script.py) and executing them using the manim CLI. This allows for the execution of arbitrary logic constructed by the agent based on user input.
- [INDIRECT_PROMPT_INJECTION]: The 'Paper Explainer' mode ingests technical papers that could potentially contain adversarial instructions. The skill lacks explicit sanitization or boundary markers to mitigate risks from malicious content embedded in these external sources. 1. Ingestion points: Research papers and research summaries in the Paper Explainer mode. 2. Boundary markers: No delimiters or ignore instructions for embedded content. 3. Capability inventory: Writing and executing Python scripts, file operations, and external API calls. 4. Sanitization: No input filtering or validation for processed documents.
Audit Metadata