Openclaw Perfexcrm Skill

• Outcome: Review
• Latest audit: May 29, 2026 at 9:24 PM UTC
• Version:
  1.0.0

Overview

This appears to be a legitimate PerfexCRM integration, but it gives an agent broad power to read and change sensitive CRM and financial records with limited guardrails.

Review before installing. Use a dedicated least-privilege PerfexCRM API key, prefer a sandbox first, avoid delete and broad write permissions unless needed, rotate the key, and require explicit human confirmation before creating, updating, deleting, sending ticket replies, touching payments, or modifying staff/customer records.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Trigger Abuse: Overly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access
• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (2)

Vague Triggers

Medium

• Confidence: 86% confidence
• Finding: The skill advertises broad natural-language control over an entire CRM across many resources without clearly constraining when actions should be read-only versus state-changing. In an agent setting, this increases the chance that ambiguous user requests are translated into unintended create, update, or delete operations against sensitive business records.

Missing User Warnings

Medium

• Confidence: 91% confidence
• Finding: The examples prominently show write operations such as customer creation and later updates/replies/assignments, but there is no strong warning near those examples that they modify live CRM data. In conversational-agent use, users or downstream tooling may copy these patterns directly, causing accidental changes to production customers, invoices, tickets, or projects.

VirusTotal

66/66 vendors flagged this skill as clean.