command-dispatch

Pass

Audited by Gen Agent Trust Hub on May 1, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill provides the ability to execute system-level commands, including sensitive operations such as system restarts (/restart) and clearing caches (/clear_cache).- [PROMPT_INJECTION]: The skill processes untrusted natural language from users to determine which commands to run and to extract arguments for execution, creating a surface for indirect prompt injection.
  • Ingestion points: Natural language descriptions provided by the user (e.g., "Sync sites", "Show subscriptions").
  • Boundary markers: Not present. The instructions do not define delimiters for user input or include warnings to ignore instructions within the data.
  • Capability inventory: Access to the run_slash_command tool, which can execute system and plugin functions.
  • Sanitization: Not present. There is no instruction for the agent to validate or sanitize user-provided arguments before executing them.
Audit Metadata
Risk Level
SAFE
Analyzed
May 1, 2026, 06:56 AM
Security Audit — agent-trust-hub — command-dispatch