skills/ml8s/liki/liki-bazi/Gen Agent Trust Hub

liki-bazi

Pass

Audited by Gen Agent Trust Hub on Sep 14, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill connects to an external RPC endpoint at https://liki.hk/jsonrpc to perform astrological calculations and check for version updates. These operations are functional for the skill's purpose and target the vendor's own infrastructure.
  • [COMMAND_EXECUTION]: The skill utilizes a local Python script (tools/agent_cli.py) and a Windows command script (tools/agent_cli.cmd) as a bridge between the AI agent and its internal logic. The adapter is securely designed with a hardcoded whitelist to ensure the agent can only invoke specific, predefined functions, effectively mitigating potential code injection or command abuse.
  • [DATA_INTEGRITY]: The skill implements a sophisticated canonical SHA-256 digest system (pan_integrity.py and pan_schema.py) to verify that the astrological data returned by the RPC is not tampered with or hallucinated by the model before being processed by subsequent analysis layers.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 14, 2026, 05:01 PM
Security Audit — agent-trust-hub — liki-bazi