xy-archive
Pass
Audited by Gen Agent Trust Hub on Sep 25, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to execute a hardcoded Python command (
python3 -c 'import datetime as d; print(d.datetime.now().astimezone().isoformat("T","seconds"))') to ensure standardized ISO 8601 timestamps with timezones for file metadata. This is a functional, single-purpose command with no user input interpolation.- [INDIRECT_PROMPT_INJECTION]: The skill functions as a data ingestion and storage layer, creating markdown files that are intended to be read back into the agent context in future sessions. This creates a surface for indirect prompt injection if malicious content is stored in the archives. However, the skill treats this data as text and provides guidance on securing these files (e.g., using .gitignore).- [SAFE]: The skill explicitly warns users that archives are stored in plain text and may contain sensitive business data. It provides user-controlled mechanisms for setting storage locations and managing file paths without performing hidden network operations or unauthorized file access.
Audit Metadata