export-knowledge
Pass
Audited by Gen Agent Trust Hub on Nov 11, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes session data (untrusted input) and writes it to the local filesystem. \n
- Ingestion points: Session history and user-provided technical findings (SKILL.md). \n
- Boundary markers: The skill structures data into specific Markdown headers and frontmatter, providing a template for output. \n
- Capability inventory: File system write access for Markdown files and a JSON index. \n
- Sanitization: Includes instructions to mask personal information and company names before recording knowledge.\n- [COMMAND_EXECUTION]: The skill uses the
Get-Datecommand to retrieve the current time for metadata purposes. This is a benign use of a standard system utility.
Audit Metadata