cinemacc-subtitle
Pass
Audited by Gen Agent Trust Hub on Aug 30, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests and processes external subtitle files (SRT), which creates a potential surface for indirect prompt injection where malicious instructions could be embedded in the subtitle text. Ingestion points: Stage 1 of SKILL.md and references/source-acquisition.md involve downloading subtitles from external providers. Boundary markers: The instructions do not mandate specific delimiters or safety headers when the agent reads external file content. Capability inventory: File system access, shell execution of local Python scripts, and web browsing. Sanitization: The skill provides a Python tool (srt_tools.py) that performs structural validation, linting, and quality assurance checks to identify unexpected content patterns.
- [DYNAMIC_EXECUTION]: The file scripts/test_srt_tools.py utilizes dynamic loading to execute the skill's main script for automated testing. Evidence: The script uses importlib.util.module_from_spec and SPEC.loader.exec_module to load and run scripts/srt_tools.py.
Audit Metadata