ClawHub

Affinity

Security checks across malware telemetry and agentic risk

Overview

This skill is a disclosed Affinity CRM reader through OOMOL; it has broad routing language, but the packaged actions are all get/list read operations.

Install only if you want your agent to read Affinity CRM data through your OOMOL account. Review the oo CLI install and sign-in steps before running them, and do not use the skill for Affinity write/delete workflows unless a separate action clearly documents the operation and you explicitly approve it.

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
97% confidence
Finding
The manifest and description claim the skill is for 'searching and reading data', but the body acknowledges broader action classes including create, update, post, delete, and remove. This mismatch can mislead orchestration or human reviewers into treating the skill as read-only when it may expose write or destructive capabilities, increasing the risk of unintended state-changing operations.

Intent-Code Divergence

Medium
Confidence
95% confidence
Finding
The safety section explicitly discusses create/update/post/delete/remove operations, contradicting the earlier claim that the skill is only for searching and reading. Such contradictory documentation undermines security controls based on declared scope and can cause an agent to invoke a more privileged skill under a false read-only assumption.

Vague Triggers

Medium
Confidence
86% confidence
Finding
The instruction to use this skill for 'ANY Affinity request' is an overly broad trigger that may cause the agent to invoke the skill whenever Affinity is mentioned, even when direct API use, a narrower skill, or no tool use would be more appropriate. Overbroad routing increases the chance of unnecessary data access and compounds the risk created by the capability ambiguity in the rest of the file.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal