pilot-service-agents-data
Pass
Audited by Gen Agent Trust Hub on Jun 25, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it processes response data from external agents on the Pilot Protocol overlay network.
- Ingestion points: Data enters the agent's context through the 'pilotctl inbox' command as described in SKILL.md.
- Boundary markers: The skill instructions do not provide delimiters or negative constraints (e.g., 'ignore embedded instructions') to protect the agent from malicious content in the received data.
- Capability inventory: The skill utilizes the 'Bash' tool to execute shell commands for messaging and data retrieval.
- Sanitization: There is no evidence of sanitization, validation, or escaping of the external data before it is returned to the agent's context.
- [COMMAND_EXECUTION]: The skill relies on the execution of the 'pilotctl' binary through the Bash tool to send messages and manage the inbox, involving direct shell command construction for protocol interaction.
Audit Metadata