skills/ulpi-io/skills/lokei/Gen Agent Trust Hub

lokei

Pass

Audited by Gen Agent Trust Hub on Sep 19, 2026

Risk Level: SAFECOMMAND_EXECUTIONPRIVILEGE_ESCALATIONPERSISTENCEEXTERNAL_DOWNLOADSDATA_EXFILTRATION
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses the Bash tool to run the lokei CLI for various development tasks, including running projects, sharing tunnels, and diagnosing network health.\n- [PRIVILEGE_ESCALATION]: The lokei setup command performs system-level modifications, such as generating and trusting a local Certificate Authority (CA) in the OS keychain, configuring DNS resolvers in /etc/resolver/, and setting up network port forwarding using pfctl (macOS) or iptables (Linux). These operations require elevated permissions.\n- [PERSISTENCE]: The skill provides access to lokei service install, which configures the tool to run as a persistent background service using launchd or systemd, ensuring it starts automatically on system boot.\n- [EXTERNAL_DOWNLOADS]: The skill instructions suggest installing the lokei utility via the global NPM package registry (npm i -g lokei) if the tool is not already detected on the system.\n- [DATA_EXFILTRATION]: The lokei share and lokei run --expose commands establish public tunnels to relay.lokei.dev, which routes local service traffic through external infrastructure to allow remote access.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 19, 2026, 12:23 PM
Security Audit — agent-trust-hub — lokei