short-drama-storyboard
Pass
Audited by Gen Agent Trust Hub on Aug 19, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses a local Python toolset (
scripts/storyboard_check.pyandscripts/selftest.py) to perform arithmetic and structural validation of storyboard data. These scripts are invoked via the command line to ensure that shot durations sum correctly and that keyframe boundaries are valid. The scripts use standard Python libraries for JSON processing and regex, and they do not attempt network access or sensitive file operations. - [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it processes untrusted screenplay data which is then used to generate structured storyboards and keyframe prompts.
- Ingestion points: The skill reads
screenplay.mdandscreenplay-index.jsonl, which are external data files provided by the user or an upstream writing process. - Boundary markers: Absent; the skill instructions for processing the screenplay do not specify the use of delimiters or 'ignore' instructions for the embedded content.
- Capability inventory: The skill has the ability to write to the local file system (saving storyboard, shot, and keyframe files) and executes local Python scripts as part of its workflow.
- Sanitization: There is no evidence of sanitization for natural language instructions that might be embedded in the screenplay text (e.g., in dialogue or action lines).
Audit Metadata