touying-author
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill allows the agent to refactor and analyze Typst source code, creating a surface for processing potentially untrusted data.
- Ingestion points: Typst source files (.typ) provided for authoring, refactoring, or troubleshooting.
- Boundary markers: No specific delimiters or instructions to ignore embedded commands are provided for processing input code.
- Capability inventory: The agent is empowered to perform file system operations (read/write) and execute shell commands related to the Typst compiler.
- Sanitization: No mechanisms for sanitizing or validating document content are included in the skill's instructions.
- [COMMAND_EXECUTION]: The skill documentation describes the use of command-line tools for presentation metadata generation and documentation management.
- Evidence: Specific instructions for using typst query to generate .pdfpc files and npm commands for managing library documentation are present in the documentation files.
Audit Metadata