tgw-asset-export
Pass
Audited by Gen Agent Trust Hub on Jun 29, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes local Python scripts, specifically
scripts/export_assets.pyandscripts/validate_asset_cards.py, to scan directories and package files. This is standard behavior for the tool's stated purpose of asset management. - [PROMPT_INJECTION]: The skill contains a surface for indirect prompt injection due to its processing of file metadata. Ingestion points: Markdown file frontmatter processed in
scripts/export_assets.py. Boundary markers: None used when writing metadata to summary files. Capability inventory: Local file reading, copying, and archiving. Sanitization: Metadata values are interpolated intoINDEX.csv,MANIFEST.md, andEXPORT_SUMMARY.mdwithout validation or escaping, though filenames are sanitized for the file system. - [SAFE]: A minor string concatenation technique is used in the Python script to check for a legacy metadata key (
tgravity_asset). This appears to be for backward compatibility and does not pose a security risk to the user.
Audit Metadata