papercut-qa
Pass
Audited by Gen Agent Trust Hub on Oct 13, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes several project-specific scripts and tools, including 'papercut-setup', 'papercut-build', and '$PAPERCUT_LIB/tools/render.sh'. These are used to scaffold the project, build scenes, and perform final video rendering.
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and act upon data from multiple external files within the project directory, such as 'style-guide.md', 'frame-breakdown.md', and 'SERIES.md'. These files influence the agent's QA logic and the parameters it passes to execution tools.
- Ingestion points: Reads content from 'style-guide.md', 'frame-breakdown.md', 'SERIES.md', and 'motion-principles.md'.
- Boundary markers: The instructions do not define delimiters or provide 'ignore embedded instructions' warnings for the data being read.
- Capability inventory: The skill can execute shell scripts ('qa.sh', 'render.sh') and build tools ('papercut-build', 'papercut-voice').
- Sanitization: There is no mention of sanitizing or validating the content of the project files before the agent processes them or uses them to generate command arguments.
Audit Metadata