skills/rheinmir/setup/medic/Gen Agent Trust Hub

medic

Pass

Audited by Gen Agent Trust Hub on Jul 3, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill is designed to execute local scripts and binaries, specifically python3 fdk/tools/medic.py or medic from ~/.local/bin/. These commands are run with arguments derived from user input.
  • [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface (Category 8) because it instructs the agent to determine the execution scope based on user descriptions. If the agent does not strictly map user input to known safe keywords, it could facilitate command injection.
  • Ingestion points: The [phạm vi] (scope) parameter in the execution step is derived from free-text user descriptions.
  • Boundary markers: There are no explicit delimiters or instructions to ignore malicious characters embedded within the user-provided scope.
  • Capability inventory: The skill utilizes shell execution via python3 or direct binary invocation.
  • Sanitization: No explicit sanitization or validation logic is defined for the user-supplied scope before it is interpolated into the command string.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 3, 2026, 04:59 PM
Security Audit — agent-trust-hub — medic