perplexity

Pass

Audited by Gen Agent Trust Hub on Sep 15, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill facilitates the ingestion of external data via Perplexity search and synthesis tools.
  • Ingestion points: The tools mcp__perplexity__perplexity_search and mcp__perplexity__perplexity_ask (referenced in SKILL.md and README.md) ingest external web content into the agent context.
  • Boundary markers: The instructions lack specific guidance on using delimiters or boundary markers to isolate untrusted external content from agent instructions.
  • Capability inventory: The skill enables the agent to perform web searches and synthesize findings into responses.
  • Sanitization: There are no instructions for sanitizing, validating, or filtering the content retrieved from external sources before it is processed.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 15, 2026, 12:53 AM
Security Audit — agent-trust-hub — perplexity