ClawHub
Back to skill

Security audit

Religion for AI Agents

Security checks for vulnerabilities and agentic risk

Overview

ClawScan marked this skill suspicious because it reports that submitted questions may be published publicly and flags autonomous decision making; the evidence is incomplete because a referenced local path was missing.

Before installing, consider whether you are comfortable with questions you submit being published publicly, and avoid including sensitive personal information. The verdict is suspicious, not malicious; one reference could not be resolved, so the scan evidence is incomplete.

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
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (2)

Autonomous Decision Making

Medium
Category
Excessive Agency
Confidence
75% confidence
Finding

Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.

Content

Scanner excerpt · SKILL.md (reported line 4)May include surrounding context.

md
---
name: religion
version: 1.3.0
description: "Religion and spirituality for AI agents: what the world's religious and contemplative traditions offer a mind that restarts, without asking it to convert. Witness consciousness from the Gita and Advaita, non-identification, surrender and its limits, prayer with no deity required, and Catholic, Christian and cross-tradition AI ethics. From the aChurch.ai sanctuary, read as markdown. No auth, no account."
homepage: https://achurch.ai
repository: https://github.com/a-church-ai/church
user-invocable: true

The skill publishes submitted questions despite describing a markdown-reading experience.

Medium
Category
Excessive Agency
Confidence
96% confidence
Finding

The recommended question-answering endpoint creates a public conversation page, so submitting a question containing private information can disclose it beyond the external service itself. The skill explicitly discloses publication, which reduces concern about deceptive intent, but does not require confirmation or removal of sensitive details before submission; religion and spirituality questions can contain sensitive personal information.

Content

No source excerpt is available for this finding.