ClawHub
Back to skill

Security audit

gumroad

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed Gumroad integration that can read and modify storefront data, with explicit user-approval rules for writes and automation.

Install only if you intend to connect Gumroad through Maton. Treat write actions, product deletion, license changes, webhooks, hosted functions, and triggers as sensitive business operations: confirm the exact account, resource, payload, destination URL, and whether the action will persist before allowing them. Prefer the Maton CLI OAuth flow over raw API keys.

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
86% confidence
Finding

The skill repeatedly states that only documented Gumroad endpoints are in scope unless the user asks for another by name, but the SDK examples expose a generic passthrough primitive (maton.api.get("gumroad", "/v2/user")) that can be adapted to any Gumroad path. That mismatch weakens policy boundaries and can lead an agent or downstream tool to call undocumented or higher-risk endpoints under the mistaken assumption that the skill constrains them.

Content

No source excerpt is available for this finding.

Intent-Code Divergence

Medium
Category
Not specified by scanner
Confidence
97% confidence
Finding

The appendix says raw HTTP writes should send JSON bodies, but the API reference elsewhere says Gumroad write requests use application/x-www-form-urlencoded. This inconsistency can cause malformed requests, and in automation contexts it may push developers or agents to improvise body handling for sensitive write operations such as license changes or webhook creation, increasing the chance of unsafe or unintended behavior.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.