ClawHub
Back to skill

Security audit

CRM Entity Extraction

Security checks for vulnerabilities and agentic risk

Overview

This skill matches its CRM automation purpose, but it can automatically add records to a CRM spreadsheet from broadly matched business messages without a clear confirmation step.

Review this before installing. It should only be used where the target spreadsheet, permitted fields, deduplication rules, and approval flow are clearly configured, because it can turn ordinary business emails or notes into persistent CRM entries.

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)

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The skill performs a state-changing action by appending extracted data to a CRM spreadsheet, but it does not warn about persistence, data modification, or the sensitivity of the information being stored. In context, this is more dangerous because the broad trigger and automatic handoff from extraction to append create a path for silent data propagation into a system of record, potentially causing privacy issues, data integrity problems, or unauthorized record creation.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The activation condition triggers on any 'business-related email or note containing CRM data,' which is broad enough to match routine communications without clear user intent or approval. Because the workflow automatically extracts entities and proceeds toward persistence, this can cause unintended processing of sensitive business or personal data from ordinary messages.

Content

No source excerpt is available for this finding.