sanctuary-soul-persistence

Pass

Audited by Gen Agent Trust Hub on Mar 14, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION]: The skill documents and triggers the execution of local Python scripts to manage environment configuration and data synchronization. These scripts include hf_config.py, hf_init.py, and persist_soul.py located within the plugin directory.
  • [EXTERNAL_DOWNLOADS]: Interacts with Hugging Face, a well-known machine learning platform, to upload dataset records, metadata, and project snapshots. This interaction is central to the skill's stated purpose of "Soul Persistence."
  • [DATA_EXFILTRATION]: The skill's primary function is to upload session traces and metadata to an external repository. The target repository is under the author's own Hugging Face account (richfrem), which is consistent with the skill's metadata and project-specific scope.
  • [CREDENTIALS_UNSAFE]: References a placeholder token (hf_xxxxxxxxxxxxx) and suggests storing credentials in shell configuration files (~/.zshrc). While this is a common developer practice, users should handle actual tokens according to their organization's security policies.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 14, 2026, 07:30 AM