master-tsongkhapa

Pass

Audited by Gen Agent Trust Hub on May 10, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill implements a 'HARD-GATE' mechanism and strict 'Iron Rules' to prevent the agent from making unverified doctrinal claims, engaging in sectarian judgment, or revealing restricted esoteric instructions (e.g., Tantric rituals).
  • [SAFE]: The skill utilizes local Python scripts (scripts/cite.py and scripts/query.py) to verify citations and query knowledge bases. This is a standard and safe pattern for ensuring factual accuracy and grounding in specialized knowledge agents.
  • [SAFE]: References and citations are directed towards established digital archives such as BDRC (Buddhist Digital Resource Center) and 84000.co, which are well-known and reputable repositories for Buddhist studies.
  • [SAFE]: No obfuscation, data exfiltration patterns, privilege escalation, or unauthorized credential access were detected in the provided files. The personality guidelines in references/voice.md and the fidelity tests in tests/fidelity.jsonl further reinforce the agent's safety and alignment with its educational purpose.
Audit Metadata
Risk Level
SAFE
Analyzed
May 10, 2026, 10:28 AM