extracting-keywords
Pass
Audited by Gen Agent Trust Hub on Apr 20, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill performs legitimate text analysis tasks using the well-known and open-source 'yake' library. No malicious patterns, such as obfuscation, hidden instructions, or credential harvesting, were identified in the source or instructions.
- [EXTERNAL_DOWNLOADS]: The installation section provides instructions to fetch 'yake', 'jellyfish', 'segtok', and 'regex' packages from official package registries via 'uv'. These are standard, reputable dependencies for natural language processing tasks.
- [COMMAND_EXECUTION]: The skill uses shell commands for routine workspace management, including the creation of a Python virtual environment and the copying of local assets to the library path to enhance functionality. All commands are transparent and limited to the agent's environment.
- [DATA_EXFILTRATION]: Data operations are confined to local file system access for reading input documents and writing extracted keywords. No network operations targeting external domains or exfiltration patterns were detected.
Audit Metadata