ClawHub

AI Compliance

Security checks across malware telemetry and agentic risk

Overview

This documentation-only compliance skill is mostly purpose-aligned, but it bundles company-specific security and compliance telemetry that should not be reused as generic guidance.

Review before installing. Use it only if the fi.com-specific content is intended for your organization and authorized for this context; otherwise remove or rewrite those sections as placeholders. Do not let an agent inspect raw prompts, DLP logs, webhook_events, or run privileged remediation commands unless a human has approved scope, redaction, and local paths.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • 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 (3)

Description-Behavior Mismatch

High
Confidence
98% confidence
Finding
This reference file is supposed to be a generic financial-services regulatory overlay, but it embeds a tenant-specific incident claim naming a company and asserting detected volumes of sensitive financial and investment-strategy events. That creates both a data leakage risk and an integrity risk: downstream agents may treat the claim as verified internal telemetry and take or recommend real-world escalation actions based on untrusted static content.

Vague Triggers

Medium
Confidence
90% confidence
Finding
The trigger scope includes a broad catch-all that activates on nearly any request to evaluate an AI tool or use case for compliance or risk. Overly broad routing can cause the skill to engage in unintended contexts, increasing the chance of applying regulatory guidance where it is incomplete, mismatched, or inappropriately authoritative.

Missing User Warnings

Medium
Confidence
94% confidence
Finding
The markdown discloses purportedly detected MNPI-related events and investment-strategy activity without any handling caveats, minimization, or provenance warning. In a compliance skill, this is especially dangerous because the content looks authoritative and could cause an agent to surface, repeat, or operationalize sensitive tenant data, amplifying confidential-information exposure and potentially triggering inappropriate legal/compliance actions.

VirusTotal

62/62 vendors flagged this skill as clean.

View on VirusTotal