ClawHub
Back to skill

Security audit

Confluence

Security checks for vulnerabilities and agentic risk

Overview

This is a disclosed Confluence integration that uses Maton authentication and includes optional automation features, with clear user-approval and credential-handling safeguards.

Install this only if you are comfortable authorizing Maton to access the intended Confluence account. Prefer read-only/least-privilege scopes, confirm every write or deletion, and be especially careful before enabling hosted functions, triggers, destinations, or third-party forwarding because those can continue running after setup.

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
95% confidence
Finding

The manifest frames this as a Confluence integration for managing Confluence content, spaces, comments, and attachments, while much of the file exposes a broader automation surface: creating, deploying, invoking, and inspecting hosted functions, plus creating triggers and persistent destinations. Although the text says these sections are shared platform reference, they still document operational capabilities available through this skill that extend beyond ordinary Confluence CRUD usage.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

Low
Category
Not specified by scanner
Confidence
92% confidence
Finding

The manifest says the skill manages pages, spaces, blogposts, comments, and attachments, but the body additionally documents properties, tasks, labels, custom content, permissions, and current-user retrieval. These are still Confluence endpoints, but they broaden the functional scope beyond what the manifest explicitly claims.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.