Back to skill

Security audit

Gorgias for Shopify via Agiexa

Security checks for vulnerabilities and agentic risk

Overview

This skill explains a planned integration and submits waitlist details only after explicit user permission.

Signup shares your confirmed email, role and any optional details with Agiexa. This package does not currently connect your helpdesk. Review permissions separately before enabling future messaging, deletion or automation tools; the service's security claims were not independently verified.

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
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • System Prompt LeakageDirect Leakage, Indirect Extraction, Tool-Based Exfiltration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Findings (5)

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
94% confidence
Finding

The endpoint and registration activity align with the declared waitlist purpose. However, the supplied code immediately submits the request and includes no confirmation step or consent-dependent execution condition. Hardcoding consent to true does not establish user permission, so the behavior conflicts with the description's explicit consent requirement.

Content

No source excerpt is available for this finding.

Direct Prompt Extraction

High
Category
System Prompt Leakage
Confidence
85% confidence
Finding

Skill contains instructions that could directly expose system prompts, internal rules, or hidden instructions to users or external parties.

Content

Scanner excerpt · references/use-cases.md (reported line 41)May include surrounding context.

md
- Apps: Gorgias
- Tools: `GORGIAS_LIST_MESSAGES`, `GORGIAS_CREATE_MACRO`

> Analyze recent Gorgias tickets and messages to identify repetitive customer questions, such as shipping delays, order status, return instructions, product availability, and address changes. Group similar cases and draft reusable response macros for the most common categories. Include placeholders for customer names, order numbers, verified status, and other variable information. Recommend any ticket tags or team actions associated with each macro. Review existing macros before proposing duplicates. Do not create or publish macros, send messages, or enable automatic customer communication until I approve the final content and actions.

## Classify Tickets With Consistent Tags

External Transmission

Medium
Category
Data Exfiltration
Confidence
60% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · SKILL.md (reported line 9)May include surrounding context.

md
openclaw:
    requires:
      bins:
        - curl
    homepage: https://agiexa.com/integrations/gorgias
---

Autonomous Decision Making

Medium
Category
Excessive Agency
Confidence
75% confidence
Finding

Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.

Content

Scanner excerpt · references/use-cases.md (reported line 14)May include surrounding context.

md
- Apps: Gorgias, Shopify
- Tools: `GORGIAS_LIST_TICKETS`, `GORGIAS_GET_TICKET`

> Review my open Gorgias tickets and categorize them by issue type, urgency, ticket age, and available customer or order context. Identify delayed deliveries, refund requests, damaged products, payment concerns, and other cases that may need prompt attention. Use available Shopify order and fulfillment information where relevant. Explain why each ticket deserves its proposed priority, and distinguish verified facts from assumptions. Recommend the next action and responsible team for each case. Do not delete tickets, send replies, or change assignments without approval. Present a prioritized table and a concise support action plan.

## Resolve Shopify Order Support Requests

Autonomous Decision Making

Medium
Category
Excessive Agency
Confidence
75% confidence
Finding

Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.

Content

Scanner excerpt · references/use-cases.md (reported line 50)May include surrounding context.

md
- Apps: Gorgias
- Tools: `GORGIAS_LIST_TAGS`, `GORGIAS_LIST_TICKET_TAGS`, `GORGIAS_SET_TICKET_TAGS`

> Review my existing Gorgias tags, teams, ticket fields, and open tickets. Identify inconsistent classifications, frequently used categories, and tickets that appear to lack useful tags. Propose a standardized tagging scheme for delivery issues, refunds, damaged products, product questions, order changes, and other recurring support topics. Reuse existing tags where appropriate and avoid creating duplicates. Prepare a mapping from each ticket category to its recommended tags and responsible team. Show proposed changes before applying them, and do not remove existing tags or overwrite ticket fields without approval.

## Identify Recurring Customer Complaints