send-to-minion
Pass
Audited by Gen Agent Trust Hub on Aug 26, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill facilitates the delegation of tasks to a single sub-agent using native platform tools (Codex or Claude Code). It includes explicit instructions to prevent nested delegation, maintain parent supervision, and ensure that the user's original goals and constraints are preserved. No malicious patterns, data exfiltration, or obfuscation were detected. The use of sub-agent tools is an intended feature of the supported runtimes and is handled here with safety constraints (e.g.,
agents.max_depth = 1).
Audit Metadata