parallel-findall
Pass
Audited by Gen Agent Trust Hub on Jun 27, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill utilizes the 'parallel-cli' tool, which is a verified resource provided by the vendor 'parallel-web' as defined in the author context. The usage of 'run', 'poll', and 'enrich' commands for discovering and managing entity data is the primary and intended purpose of the skill.
- [SAFE]: Instructions for upgrading or installing 'parallel-web-tools' via pipx target official vendor packages and follow standard software management practices.
- [SAFE]: The skill manages search result data by saving it to the /tmp/ directory, which is appropriate for transient processing and adheres to standard file system usage.
- [SAFE]: To address potential risks from processing external data (Indirect Prompt Injection), the skill provides comprehensive guidance for the agent to validate and filter output from the 'findall' command. (1) Ingestion points: Entity results retrieved from the parallel-cli API; (2) Boundary markers: Not explicitly defined, though data is processed in structured formats; (3) Capability inventory: Subprocess calls via parallel-cli for data management; (4) Sanitization: Clear instructions to filter noise, validate URLs, and perform factual spot-checks on the matched entities before presentation.
Audit Metadata