aip-dynamic-caption
Pass
Audited by Gen Agent Trust Hub on Oct 30, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to execute a Python script (
caption_cues.py) provided within its own repository to synchronize caption timings with video data. This is a functional part of the skill's operation. - [EXTERNAL_DOWNLOADS]: It provides instructions to download font assets from well-known and reputable services such as Google Fonts and Fontsource.
- [INDIRECT_PROMPT_INJECTION]: The skill has an ingestion surface for untrusted data as it processes video project files to generate caption content. 1. Ingestion points: Ingests data from
index.htmlandediting-script.json(SKILL.md). 2. Boundary markers: No explicit markers or delimiters are defined in the instructions for these inputs. 3. Capability inventory: The skill can execute local Python scripts and write HTML, CSS, and JS files to the project directory. 4. Sanitization: No specific sanitization or escaping steps are described for the transcript text before it is placed into HTML templates.
Audit Metadata