researching-web

Pass

Audited by Gen Agent Trust Hub on Apr 14, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it retrieves and processes data from external, untrusted web sources.
  • Ingestion points: The skill uses the WebFetch tool to ingest content from URLs found in Perplexity AI citations in SKILL.md.
  • Boundary markers: No specific delimiters or instructions to ignore potential commands within the fetched content are implemented in the WebFetch prompts or the surrounding instructions.
  • Capability inventory: Across the skill, the agent has access to file system tools (Read, Grep, Glob) and the ability to spawn subagents (Task), which increases the potential impact if a malicious website provides instructions that the agent follows.
  • Sanitization: No sanitization or validation logic is applied to the content retrieved via WebFetch before it is synthesized into a response.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 14, 2026, 08:42 PM