moviepy
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFENO_CODEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill provides purely instructional content and code snippets for video processing using standard Python libraries (moviepy, Pillow, matplotlib). No malicious patterns, obfuscation, or unauthorized access attempts were detected.
- [INDIRECT_PROMPT_INJECTION]: The skill describes a data processing pipeline that ingests external video files and user-provided text strings. While this creates a surface for potential indirect prompt injection, the skill itself does not implement automated execution of instructions derived from these inputs. 1. Ingestion points: Video files via VideoFileClip and text labels via render_text_png in SKILL.md. 2. Boundary markers: None specified in the provided snippets. 3. Capability inventory: Local file writing for video and image output (write_videofile, img.save) in SKILL.md. 4. Sanitization: Not explicitly present in the provided templates.
- [EXTERNAL_DOWNLOADS]: The skill recommends installing standard dependencies (moviepy, Pillow, matplotlib) from official registries using the uv package manager. These are well-known and expected tools for the skill's domain.
Audit Metadata