parallel-web

Pass

Audited by Gen Agent Trust Hub on Feb 22, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSPROMPT_INJECTIONDATA_EXFILTRATION
Full Analysis
  • EXTERNAL_DOWNLOADS (LOW): The skill documentation instructs users to install parallel-web and openai. While openai is a trusted source, parallel-web is not on the approved list of trusted organizations. This is considered low risk as it is the primary package required for the skill.
  • DATA_EXFILTRATION (LOW): The skill performs network requests to api.parallel.ai. While this domain is not on the whitelist, these operations are necessary for the skill to function as a web search and extraction tool.
  • PROMPT_INJECTION (LOW): The skill exhibits an indirect prompt injection surface (Category 8) because it retrieves data from arbitrary URLs via the Search and Extract APIs. Ingestion points: Web content and excerpts returned by the api.parallel.ai endpoints. Boundary markers: Absent; the documentation does not provide instructions for the agent to use delimiters or ignore instructions found within extracted text. Capability inventory: The skill allows the agent to read and process external web content. Sanitization: No sanitization or filtering of the fetched content is described in the provided documentation.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 22, 2026, 02:45 AM