timesfm-forecasting

Pass

Audited by Gen Agent Trust Hub on Jun 21, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: Model weights (~800MB) are downloaded from the official Google Research repository on Hugging Face (google/timesfm-2.5-200m-pytorch). This is a standard operation for foundation model skills using a well-known model provider.\n- [COMMAND_EXECUTION]: The preflight checker script (scripts/check_system.py) executes system utilities like sysctl and vm_stat on macOS to verify hardware capabilities. These calls are used exclusively to ensure the local environment has enough RAM to load the model without crashing.\n- [SAFE]: The skill employs safe patterns, including dependency verification and hardware resource validation. No evidence of prompt injection, data exfiltration, or malicious obfuscation was found across the skill's scripts and configuration files.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 21, 2026, 11:09 AM