parallel-web

Pass

Audited by Gen Agent Trust Hub on Apr 21, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill requires the installation of the openai and parallel-web Python packages to interface with the Parallel Web Systems APIs. These are standard client libraries used to interact with the vendor's services.
  • [COMMAND_EXECUTION]: The script scripts/parallel_web.py provides a command-line interface for search and extraction tasks, allowing the agent to execute research operations within the environment.
  • [DATA_EXFILTRATION]: The skill transmits user queries and research objectives to the api.parallel.ai service. This data transfer is the intended purpose of the tool and uses the vendor's documented API infrastructure.
  • [PROMPT_INJECTION]: The skill has a vulnerability surface for indirect prompt injection due to its handling of untrusted data from the web.
  • Ingestion points: Web content is retrieved through the Parallel Search and Extract APIs and provided to the agent or saved to files.
  • Boundary markers: No specific delimiters or safety instructions are implemented in the script to distinguish retrieved web data from the agent's internal logic.
  • Capability inventory: The script can write output to the local filesystem (utilizing the Write tool) and performs network communication with the API provider.
  • Sanitization: No explicit sanitization, filtering, or validation of the retrieved web content is performed before it is returned to the agent context.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 21, 2026, 08:17 AM