document-qa
Pass
Audited by Gen Agent Trust Hub on Sep 10, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes a local Python script
src/scripts/check_sections.pyto automate document analysis. This tool verifies document structure, formatting redlines, and specific official document types (公文) against defined templates. - [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied document drafts (
成稿.md) as the primary data source for its validation logic. - Ingestion points: Document drafts are ingested as file arguments to the
check_sections.pyCLI script. - Boundary markers: Explicit instructions direct the agent to only report issues and avoid unauthorized rewriting of the source text.
- Capability inventory: The skill uses local shell command execution to run its internal validation scripts.
- Sanitization: No specific input sanitization or validation mechanisms for the content of the external document are detailed in the skill instructions.
Audit Metadata