parallel-web-search

Pass

Audited by Gen Agent Trust Hub on May 7, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection as it processes untrusted data from the open web.
  • Ingestion points: Web search results, titles, and content excerpts retrieved via the parallel-cli search command and saved to /tmp/ JSON files.
  • Boundary markers: Absent. The instructions do not provide delimiters or specific warnings to the agent to ignore instructions embedded within the search results.
  • Capability inventory: The skill utilizes the Bash tool and has the ability to read and write files to the /tmp directory.
  • Sanitization: Absent. The agent is instructed to read the raw JSON output and synthesize a response directly from the web content.
Audit Metadata
Risk Level
SAFE
Analyzed
May 7, 2026, 06:19 PM