pilot-service-agents-flights

Pass

Audited by Gen Agent Trust Hub on Jul 10, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill requires the pilotctl utility and uses Bash to execute network operations. Commands like pilotctl send-message and pilotctl inbox are used to interface with the Pilot Protocol backbone.
  • [PROMPT_INJECTION]: The skill processes data from various external aviation agents which constitutes a surface for indirect prompt injection. Maliciously crafted data in a flight record or weather report could attempt to override the agent's instructions during processing.
  • Ingestion points: Responses fetched via pilotctl --json inbox (SKILL.md).
  • Boundary markers: Not implemented; external data is not wrapped in delimiters or accompanied by instructions to ignore embedded commands.
  • Capability inventory: The skill has access to the Bash tool and execution of the pilotctl binary.
  • Sanitization: No data validation or escaping is described for the content received from the overlay network.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 10, 2026, 09:49 AM
Security Audit — agent-trust-hub — pilot-service-agents-flights