perplexity

Pass

Audited by Gen Agent Trust Hub on Sep 22, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied data through the query field, which is passed to the Perplexity AI service. This creates a surface for instructions embedded in search queries to influence agent behavior.
  • Ingestion points: Input JSON query parameter defined in SKILL.md.
  • Boundary markers: Absent; the instructions do not specify delimiters or system instructions to ignore embedded commands in the query.
  • Capability inventory: The skill uses the refly CLI to trigger workflows and retrieve AI responses from external services.
  • Sanitization: No validation or sanitization of the query content is described.
  • [COMMAND_EXECUTION]: The execution guide instructs the user to run multiple shell commands using the refly CLI and jq. While these commands are standard for interacting with the Refly platform, they involve executing external binaries to process data.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 22, 2026, 11:23 AM