ClawHub
Back to skill

Security audit

microsoft-excel

Security checks for vulnerabilities and agentic risk

Overview

This is a disclosed Microsoft Excel integration that uses Maton OAuth and CLI access, with clear warnings around writes, credentials, and automation.

Install this only if you want Maton to access Excel data in your Microsoft account. Before using it, confirm the account and scopes, approve every write or automation separately, keep trigger scopes narrow, and only use the non-Excel trigger references when you intentionally want cross-app automation with a named source and destination.

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
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • 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 (2)

Context-Inappropriate Capability

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

This Excel skill bundle includes detailed Linear trigger documentation for a non-Excel event source, which can mislead an agent into using capabilities outside the skill's stated purpose. Even though the text says it is shared reference material, it still provides actionable event types, parameters, and payload shapes that could cause cross-connector data handling or unintended automation if the agent treats it as in-scope.

Content

No source excerpt is available for this finding.

Intent-Code Divergence

Low
Category
Not specified by scanner
Confidence
82% confidence
Finding

The file attempts to disclaim scope expansion, but immediately provides concrete Linear event-handling guidance, including scoping, forwarding, and destination recommendations. That contradiction increases the chance an agent will operationalize the unrelated Linear integration anyway, creating confusion about authorized data flows and potentially exposing personal data from webhook payloads.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.