manim-composer

Pass

Audited by Gen Agent Trust Hub on Nov 14, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests external research data from web search and user-provided concepts, creating a standard attack surface for indirect prompt injection.
  • Ingestion points: Video concepts provided by the user and research data fetched via web search (Phase 1 and 2 in SKILL.md).
  • Boundary markers: The instructions do not specify delimiters or constraints to separate research content from the agent's core instructions.
  • Capability inventory: The skill is limited to content generation (writing scenes.md) and does not perform command execution or network exfiltration.
  • Sanitization: No explicit validation or sanitization of external data is defined.
  • [NO_CODE]: The skill consists entirely of Markdown reference files and templates. It does not contain any executable scripts, binaries, or automated installation procedures.
  • [SAFE]: The skill follows best practices for planning and documentation. Code snippets included in the reference files are provided as examples for the user rather than for direct execution by the agent.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 14, 2026, 08:42 AM