Security audit

master-atisha

Security checks across malware telemetry and agentic risk

Overview

This skill is a coherent religious teaching assistant for Atiśa/Kadam topics, with limited optional web lookup for citations and no persistence or hidden install behavior.

Installers should expect this skill to answer from an Atiśa/Kadam Buddhist perspective, often in a Chinese teaching style, and to optionally query fojin.app for source citations when local excerpts are insufficient. Review the broad trigger wording if you only want it used for explicit Atiśa or Kadam requests.

SkillSpector

By NVIDIA

Vulnerability Patterns

Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (5)

Vague Triggers

Medium

Confidence: 95% confidence
Finding: This markdown skill description applies to SQP-1. The phrase 'invoke whenever user's question touches Kadam / lamrim foundations / bodhicitta cultivation, even without explicit request' is ambiguous and broad, because 'touches' does not define clear boundaries and the listed topics are expansive enough to overlap with general religious or philosophical discussion.

Natural-Language Policy Violations

Low

Confidence: 87% confidence
Finding: SQP-3 applies to all file types. The instruction '语言：简洁、直接、慈悲恳切；常以印度比喻入题' imposes a fixed linguistic/style behavior, and the file does not offer the user any language or locale choice elsewhere despite being multilingual in description.

Natural-Language Policy Violations

Medium

Confidence: 82% confidence
Finding: The natural-language style instructions are written entirely in Chinese and specify how the skill should answer, but the file does not state that language choice is optional or user-selectable. Because SQP-3 applies to all file types, this can be treated as a locale/language policy issue when no opt-in or justification is provided.

Natural-Language Policy Violations

Low

Confidence: 87% confidence
Finding: This markdown file presents all instructional content in a single language and does not indicate that users can choose another language or that the Chinese-only format is required for a region-specific purpose. Under the stated policy, forcing a specific language without user opt-in is a natural-language policy concern.

Natural-Language Policy Violations

Medium

Confidence: 89% confidence
Finding: The entire skill specification is written to enforce a Chinese-speaking response style and provides prescribed Chinese openings, forms of address, and closing reminders, but it does not offer the user a language choice or state that Chinese is optional. Under the policy, forcing a specific language without user opt-in is a natural-language policy violation.

VirusTotal

62/62 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.