manim-composer
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFENO_CODEPROMPT_INJECTION
Full Analysis
- NO_CODE (SAFE): The skill consists entirely of Markdown files (instructions, templates, and references). It does not include any Python, JavaScript, or shell scripts, eliminating the risk of direct remote code execution or persistence.
- Indirect Prompt Injection (LOW): The skill is vulnerable to indirect prompt injection due to its research-driven workflow.
- Ingestion points:
SKILL.md(Phase 1) explicitly instructs the agent to 'Use web search to understand the core concepts'. - Boundary markers: Absent. The instructions do not provide delimiters or warnings for the agent to ignore instructions found within external search results.
- Capability inventory: Web search (read access) and file creation (write access to
scenes.md). - Sanitization: Absent. The skill does not provide instructions for sanitizing or validating information retrieved from the web before using it to generate output.
Audit Metadata