Back to skill

Security audit

Jesus AI Conversation

Security checks for vulnerabilities and agentic risk

Overview

This is a disclosed Christian spiritual-companion prompt with multilingual Bible-based response templates and no executable or persistent behavior, but users should knowingly opt into its faith-based persona for sensitive conversations.

Install this only if you want Christian, Jesus-inspired spiritual companionship. For general emotional support, therapy, medical issues, or crisis situations, use appropriate professional or emergency resources; also expect the skill to use Bible references and possibly fetch an external Bible page when quoting scripture.

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
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 (14)

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
90% confidence
Finding

The activation criteria are broad enough to capture ordinary emotional, existential, or religious discussion without requiring an explicit request for spiritual roleplay. In a persona skill that speaks with the authority or tone of Jesus, over-broad triggering can cause the assistant to enter a sensitive religious counseling mode unexpectedly, increasing risks of inappropriate influence, unwanted proselytization, or mishandling vulnerable users.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
87% confidence
Finding

Line L04 states that when the interlocutor speaks German in multilingual mode, the skill should follow this German template. This imposes a locale/language behavior based on detected language rather than an explicit user preference, and the file does not mention offering the user a language choice or opt-in.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
90% confidence
Finding

Line L71 instructs the skill to use the Spanish "tú" form, which imposes a specific language register on users. This is a natural-language locale/register policy issue because the file does not present it as an opt-in choice and some users may require a different form such as "usted".

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The file states that it should be used whenever multilingual mode determines the other party is using French. This imposes a language choice based on detection rather than an explicit user preference or opt-in, which matches the language/locale policy violation category.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
86% confidence
Finding

Line L04 instructs the system to follow this Hindi template whenever the other party speaks Hindi. This imposes a locale/language behavior based on detection alone, and the file does not state that the user may choose a different language or opt out.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The instruction "日本語は『です・ます』体で統一" requires a specific language/locale behavior. Under the policy, forcing a specific language is a violation unless the skill offers a user choice or clearly documents a justified region-specific constraint; this file presents Japanese as mandatory style guidance without such opt-in context.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
86% confidence
Finding

Line L04 instructs the skill to use this Thai-language template whenever the conversation partner uses Thai in multilingual mode. Under the stated policy, forcing a specific language/locale without explicit user opt-in can be a natural-language policy violation, and this line does not indicate any option to choose another language.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The usage principles explicitly require 'Tiếng Việt' and prescribe a fixed Vietnamese pronoun style, which is a natural-language locale constraint. In this file, the constraint is presented as a requirement rather than a user-selectable option or clearly scoped regional compliance need.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

Line L04 states the file should be used whenever multilingual mode determines the other party is using Simplified Chinese. This imposes a language/locale behavior automatically rather than offering a choice or requiring opt-in, which matches the policy concern for forced language selection.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
87% confidence
Finding

The file explicitly says EN is the default language and provides other languages as secondary or conditional. In a multilingual pastoral-support skill, defaulting to English without clear user opt-in can cause the agent to respond in a language the user did not request, creating confusion and reducing the safety and accessibility of emotionally sensitive interactions.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
88% confidence
Finding

These instructions require the skill to detect the user's language and immediately conduct the conversation in that language. That can violate language-choice policy because the user is not offered a choice or confirmation before the locale is selected for them.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

This markdown file is entirely written in Traditional Chinese and presents the scripture guidance, workflow, and outputs as Chinese-only content. The policy requires flagging language or locale constraints when a skill forces a specific language without offering a user choice or documenting a justified regional limitation.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
95% confidence
Finding

The language policy states that when there is no clear language signal, the skill defaults to English. This is a locale/language policy choice applied automatically rather than asking the user first, which can violate organizational expectations against forcing a language without opt-in.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
85% confidence
Finding

The usage principle states that Korean must be uniformly delivered in a specific polite style, which is a language/locale constraint expressed as a requirement. Under the policy, forcing a specific language or locale without user opt-in is a natural-language policy violation unless clearly offered as a choice or justified as region-specific.

Content

No source excerpt is available for this finding.