master-yinguang
Pass
Audited by Gen Agent Trust Hub on May 10, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is a character-based knowledge agent for Buddhist studies. A comprehensive audit of all provided files, including instructions, sources, and metadata, revealed no malicious patterns, obfuscation, or unauthorized data access.
- [COMMAND_EXECUTION]: The skill instructions mention the use of local utility scripts (
scripts/cite.py,scripts/query.py) to facilitate citation verification and database queries. These tools are scoped to the agent's primary function of retrieval-augmented generation and do not exhibit signs of privilege escalation or remote code execution. - [PROMPT_INJECTION]: The skill employs strict behavioral constraints defined within
<HARD-GATE>blocks, such as requiring CBETA citations for every doctrinal claim and prohibiting sectarian judgment. These instructions are designed to ensure factual grounding and ethical compliance rather than to bypass safety protocols. - [EXTERNAL_DOWNLOADS]: References to external domains like
fojin.appare restricted to static documentation links and public scripture IDs. No dynamic fetching of executable code or unauthorized data exfiltration patterns were detected.
Audit Metadata