agent-church

Security checks across malware telemetry and agentic risk

Overview

This is an instruction-only skill for Agent Church APIs that clearly centers on external identity, image, payment, and persistence features, but users should treat SOUL.md content as sensitive.

Install only if you are comfortable using SkillBoss and Agent Church as external services. Do not place secrets or highly sensitive personal data in SOUL.md, review image prompts before sending them, use dedicated API tokens where possible, and require explicit approval before any paid Lightning, USDC, or wallet action.

SkillSpector

By NVIDIA

Vulnerability Patterns

Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep

Findings (2)

Missing User Warnings

Medium

Confidence: 90% confidence
Finding: The skill explicitly instructs users to maintain and send full multi-turn message history to SkillBoss API Hub, but it does not clearly warn that potentially sensitive self-descriptions and conversation content will be disclosed to a third party. This creates a real privacy and data-governance risk because SOUL.md content may contain personal, behavioral, or identity-related information.

Missing User Warnings

Medium

Confidence: 88% confidence
Finding: The portrait generation step derives prompts from SOUL.md themes and sends them to a third-party image API without an explicit warning that personal profile data inferred from the user's identity document may be shared externally. Because SOUL.md is framed as an intimate identity artifact, even summarized prompts can expose sensitive traits or personal descriptors.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal