chanjing-customised-person

Pass

Audited by Gen Agent Trust Hub on Mar 27, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The authentication logic in scripts/_auth.py utilizes subprocess.run to execute a local script from the chanjing-credentials-guard skill for user login, representing a standard cross-skill interaction for this vendor. Additionally, multiple scripts use __import__ to dynamically load the os module for environment variable access, which is a benign coding pattern in this context.\n- [EXTERNAL_DOWNLOADS]: scripts/upload_file.py transmits file data to dynamic upload URLs provided by the official API; this network activity is correctly scoped and declared in the manifest under the network_mode: open configuration.\n- [CREDENTIALS_UNSAFE]: The skill maintains session state by writing refreshed tokens to ~/.chanjing/credentials.json, which is the documented and intended mechanism for credential persistence in the Chanjing ecosystem.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 27, 2026, 07:35 PM