krt-compound-master

Pass

Audited by Gen Agent Trust Hub on May 20, 2026

Risk Level: SAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS]: Recommends installing additional tools from the author's repository (ElZaWarudo/krt) using the npx package runner.
  • [COMMAND_EXECUTION]: Facilitates system operations through delegated roles like worker and explorer, with strict protocols for running verification commands.
  • [PROMPT_INJECTION]: The skill uses structured, goal-oriented prompts for its sub-agents, focusing on implementation consistency without bypass attempts.
  • [CREDENTIALS_UNSAFE]: Prohibits the collection of Jira credentials, relying on environment variables for API access.
  • [DATA_EXFILTRATION]: Restricts sub-agents from performing unauthorized network operations, delegating communication to specialized handoff roles.
Audit Metadata
Risk Level
SAFE
Analyzed
May 20, 2026, 07:08 AM
Security Audit — agent-trust-hub — krt-compound-master