ClawHub
Back to skill

Security audit

Monday.com

Security checks for vulnerabilities and agentic risk

Overview

This is a disclosed Monday.com automation skill that can access and change business data through Maton, but its sensitive actions are described and approval-gated.

Use this only if you trust Maton with your Monday.com account. Prefer OAuth over long-lived API keys, grant the narrowest scopes available, and carefully review any approval request for writes, deletions, hosted functions, triggers, or destinations, especially anything that forwards data to a third-party URL.

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
  • 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
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

Description-Behavior Mismatch

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The manifest and earlier body repeatedly state that Monday.com is GraphQL-only and every call is a POST to /monday/v2 (L0031, L0160, L0935). But the appendix says read-only calls can omit data and method, and suggests changing method= to PUT/DELETE, which describes behaviors this skill does not actually use for Monday.com and misstates how the integration works.

Content

No source excerpt is available for this finding.

Intent-Code Divergence

Low
Category
Not specified by scanner
Confidence
83% confidence
Finding

The manifest text says the functions/triggers sections 'add no Monday.com capability,' yet the file goes on to document creation and management of hosted functions, triggers, event watches, and persistent destinations (for example L0212-L0216, L0614-L0619, L0710-L0714, L0875-L0888). While these are platform capabilities rather than Monday API calls, the wording is stronger than the implementation and documentation support, because the skill clearly enables automation behaviors beyond direct Monday CRUD/query operations.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.