ClawHub
Back to skill

Security audit

Real Estate — Sphere-of-Influence Referral Ladder

Security checks for vulnerabilities and agentic risk

Overview

This is a plain real-estate referral-planning skill with no code, hidden actions, credential use, or persistence.

Installers should understand that this skill may be selected for realtor CRM and referral-nurture planning. It does not itself access contact databases or send messages, but users should still review any separate CRM automation or outreach actions before allowing an agent to execute them.

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
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • 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 (2)

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The frontmatter activation description uses broad phrases like wanting repeat/referral business, asking where the next deal will come from, and building nurture systems. These are common business concerns that could cause the skill to activate in loosely related conversations, leading an agent to apply the wrong workflow or override a more appropriate skill. In an autonomous routing system, this increases the chance of prompt/skill misselection and unintended behavior.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
89% confidence
Finding

The main skill body repeats activation conditions that are still ambiguous, such as building a repeat/referral pipeline or deciding nurture cadence, without defining clear thresholds or exclusion rules. Because these phrases overlap with generic sales, CRM, and marketing discussions, an orchestrating agent may invoke this skill outside its intended real-estate sphere-of-influence context, causing context drift and poor task selection.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.