custom-agent-usage

Pass

Audited by Gen Agent Trust Hub on Mar 9, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill's primary function is to facilitate the discovery and use of local custom agent files. No high-risk behaviors or malicious patterns were found.
  • [EXTERNAL_DOWNLOADS]: The provided utility script scripts/agent-finder.js utilizes the glob package, which is a standard and well-known library for file system searching.
  • [PROMPT_INJECTION]: The skill features a discovery mechanism for .agent.md files. This involves parsing file metadata to inform subagent delegation tasks. This is an intended architectural feature for agent management.
  • Ingestion points: Local .agent.md files in the workspace.
  • Boundary markers: None.
  • Capability inventory: Reading workspace files and delegating tasks via runSubagent.
  • Sanitization: None (extracts text fields for display and configuration).
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 9, 2026, 10:56 PM