ClawHub
Back to skill

Security audit

MeritLoop Architect

Security checks for vulnerabilities and agentic risk

Overview

This skill is a domain-specific product-design guide for Buddhist ritual AI concepts and does not contain executable code, hidden persistence, credential handling, or data exfiltration behavior.

Install this if you want help designing Buddhist or spiritual AI product concepts. Review outputs for cultural, doctrinal, and localization sensitivity, especially around merit scoring, temple services, and language defaults, but the skill itself does not appear to add unsafe technical authority.

A.I.G

Vulnerability Patterns
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs

SkillSpector

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (8)

Context Leakage

High
Category
Data Exfiltration
Content
<text x="650" y="608" fill="#5F3118" font-size="24" font-family="'Helvetica Neue', Arial, sans-serif">1. Schedule ritual</text>
  <text x="650" y="642" fill="#5F3118" font-size="24" font-family="'Helvetica Neue', Arial, sans-serif">2. Chant or listen</text>
  <text x="650" y="676" fill="#5F3118" font-size="24" font-family="'Helvetica Neue', Arial, sans-serif">3. Dedicate merit</text>
  <text x="930" y="608" fill="#5F3118" font-size="24" font-family="'Helvetica Neue', Arial, sans-serif">4. Log session locally</text>
  <text x="930" y="642" fill="#5F3118" font-size="24" font-family="'Helvetica Neue', Arial, sans-serif">5. Update practice score</text>
  <text x="930" y="676" fill="#5F3118" font-size="24" font-family="'Helvetica Neue', Arial, sans-serif">6. Nightly spoken summary</text>
Confidence
85% confidence
Finding
Code or instructions that leak agent conversation context to external services, potentially exposing sensitive user interactions.

Vague Triggers

Medium
Confidence
89% confidence
Finding
The skill’s activation criteria are broad and can match generic product ideation requests, causing the agent to invoke this domain-specific skill when the user did not intend a Buddhist or spiritual framing. That can steer outputs toward sensitive religious-product guidance, monetization, and ritual design in contexts where a general product skill would be more appropriate, increasing misrouting and policy-boundary risk.

Natural-Language Policy Violations

Medium
Confidence
86% confidence
Finding
The SVG text explicitly states 'English as the default UI, then adapt to each local system and ritual context,' which sets a fixed default language before any user choice. This can violate language/locale policy guidance because it imposes a specific language rather than offering an explicit opt-in or neutral initial selection.

Natural-Language Policy Violations

Medium
Confidence
93% confidence
Finding
This SVG includes natural-language product text that specifies 'English-first UI.' The policy allows locale constraints only when users are given a choice or the restriction is clearly justified; this wording suggests a default language preference without opt-in or explanation.

Vague Triggers

Medium
Confidence
97% confidence
Finding
The trigger guidance says to use this file when the user wants 'naming help or a product gap review,' which is a broad natural-language condition without clear scope, examples, or exclusions. Because those phrases overlap with common requests and do not define boundaries for this skill, the file could be invoked in unintended contexts.

Natural-Language Policy Violations

Medium
Confidence
88% confidence
Finding
The text presents 'Chinese first, multilingual later' as a default regionalization option, which can imply a language or locale preference without explicit user choice. Under the policy, forcing a specific language or locale without opt-in is a natural-language policy concern unless clearly justified as region-specific.

Natural-Language Policy Violations

Low
Confidence
83% confidence
Finding
This file contains user-facing natural-language content in a specific non-English locale on line 45. Under the policy rule for language or locale constraints, presenting fixed-language content without opt-in or justification can be considered a locale-policy issue.

Natural-Language Policy Violations

Low
Confidence
81% confidence
Finding
The SVG includes the phrase "English default · multilingual local deployment," which indicates a forced default language choice in natural-language content. The file shows multilingual support, but it does not state that users can choose their preferred language or opt in to English as the default.

Static analysis

No suspicious patterns detected.