Security audit

Brand Protection Ebay

Security checks for vulnerabilities and agentic risk

Overview

The skill is not overtly malicious, but it is advertised as an eBay VeRO brand-protection toolkit while its scripts mainly generate Amazon Brand Registry and ASIN-based guidance.

Review carefully before installing. Do not rely on this as an eBay VeRO enforcement tool unless the publisher fixes the Amazon/eBay mismatch, aligns the CLI commands, and pins the installation source. Treat generated legal or complaint text as draft material that must be checked before submitting evidence to any marketplace.

A.I.G

Vulnerability Patterns

Insecure DependenciesIntroduces malicious components through unsafe dependency sources
Insecure Skill Coding PracticesFinds exploitable flaws such as hardcoded secrets or command injection
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

Findings (2)

T08 · Insecure Dependencies

Warning

Location: SKILL.md:17
Finding: Unpinned npm Package Execution During Global Installation
Content: View full analysis
Remediation: View remediation
add ... ``` 2. Pin the external Skill source to a reviewed release tag or immutable commit hash. 3. Publish and verify cryptographic checksums or signatures for released Skill artifacts. 4. Use a lockfile where the installation workflow supports one. 5. Avoid global installation unless it is operationally necessary. 6. Run installation under a non-administrative account and in an isolated environment. 7. Document the expected package publisher, repository, version, and checksum so users can validate the source before execution. ]]>

T09 · Insecure Skill Coding Practices

Note

Location: SKILL.md:4
Finding: eBay-Branded Skill Produces Amazon-Specific Enforcement Guidance
Content: View full analysis
Remediation: View remediation

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 (13)

Tp4

High

Category: MCP Tool Poisoning
Confidence: 95% confidence
Finding: This appears to market a broad eBay brand-protection capability while the underlying behavior is limited to manual document/template generation and may even target Amazon concepts instead. Such mismatch is dangerous because users may assume active detection, price monitoring, or infringement analysis is occurring when it is not, delaying response to counterfeits or unauthorized sellers.
Content: No source excerpt is available for this finding.

Tp4

High

Category: MCP Tool Poisoning
Confidence: 95% confidence
Finding: This appears to market a broad eBay brand-protection capability while the underlying behavior is limited to manual document/template generation and may even target Amazon concepts instead. Such mismatch is dangerous because users may assume active detection, price monitoring, or infringement analysis is occurring when it is not, delaying response to counterfeits or unauthorized sellers.
Content: No source excerpt is available for this finding.

Description-Behavior Mismatch

High

Category: Not specified by scanner
Confidence: 97% confidence
Finding: The core data model is built around Amazon ASINs rather than eBay item identifiers, meaning detections and reports are structurally tied to the wrong marketplace. For a security-sensitive enforcement tool, this can produce misleading results, incorrect evidence packages, and failed or improper takedown actions.
Content: No source excerpt is available for this finding.

Description-Behavior Mismatch

High

Category: Not specified by scanner
Confidence: 99% confidence
Finding: The code generates Amazon Brand Registry, counterfeit, and complaint workflows even though the skill is advertised as an eBay toolkit. In a brand-protection context, this platform mismatch can cause users to take the wrong enforcement actions, submit complaints to the wrong service, and mishandle evidence collection or legal escalation, making the skill operationally unsafe.
Content: No source excerpt is available for this finding.

Rp1

Medium

Category: MCP Rug Pull
Confidence: 89% confidence
Finding: The installation command uses npx skills without pinning an exact version, which allows whatever version is current at execution time to run. This creates a supply-chain risk: a compromised or breaking upstream release could execute unexpected code during installation in a user environment.
Content: No source excerpt is available for this finding.

Intent-Code Divergence

Medium

Category: Not specified by scanner
Confidence: 94% confidence
Finding: The header advertises image theft detection and risk evaluation features that are not actually implemented, creating a false sense of coverage. In a brand-protection workflow, users may rely on the tool to detect infringements it never checks, causing missed abuse and delayed response.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium

Category: Not specified by scanner
Confidence: 93% confidence
Finding: The CLI chooses English unless --zh is provided, but the skill also contains a format_report(..., lang="en") API that supports a forced locale and embeds substantial Chinese-language output paths throughout the file. This creates a built-in language constraint behavior without any explicit user-choice mechanism beyond a hardcoded flag, which can violate language/locale policy when integrated into broader agent flows.
Content: No source excerpt is available for this finding.

Description-Behavior Mismatch

Medium

Category: Not specified by scanner
Confidence: 97% confidence
Finding: The manifest frames the skill as focused on eBay brand protection, but this file's core outputs are complaint and enforcement templates explicitly targeting Amazon Brand Registry and Amazon seller workflows. That is a substantive behavior mismatch, not just an implementation detail, because the skill is providing platform-specific operational content for a different marketplace.
Content: No source excerpt is available for this finding.

Description-Behavior Mismatch

Medium

Category: Not specified by scanner
Confidence: 95% confidence
Finding: The manifest advertises an eBay toolkit with VeRO-oriented capabilities, but this function produces a test-buy procedure centered on Amazon buyer accounts, Brand Registry submission, and Amazon review periods. That content exceeds the declared marketplace scope and changes the practical use of the skill.
Content: No source excerpt is available for this finding.

Description-Behavior Mismatch

Medium

Category: Not specified by scanner
Confidence: 90% confidence
Finding: Although MAP monitoring is mentioned in the manifest, this template ties enforcement to Amazon-specific reporting and reseller-program consequences instead of eBay processes. That makes the implemented behavior materially different from an eBay-focused toolkit as described.
Content: No source excerpt is available for this finding.

Description-Behavior Mismatch

Medium

Category: Not specified by scanner
Confidence: 96% confidence
Finding: This localized template still directs users to Amazon Brand Registry and describes Amazon infringement submission steps, which conflicts with the manifest's eBay-specific positioning. The mismatch persists across locales, indicating the implemented scope is broader and different than claimed.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium

Category: Not specified by scanner
Confidence: 95% confidence
Finding: This file includes dedicated _zh template generators and CLI logic that switches to Chinese output via a flag, but the content does not explain any region-specific requirement or user-facing locale policy. Because the skill embeds a forced locale variant rather than offering a documented language selection policy, it creates a language/locale policy concern under the stated rule.
Content: No source excerpt is available for this finding.

Description-Behavior Mismatch

Medium

Category: Not specified by scanner
Confidence: 94% confidence
Finding: The localized test-buy workflow instructs use of Amazon buyer accounts and Brand Registry complaint submission, which is inconsistent with an eBay-only brand-protection toolkit. This is a semantic scope mismatch at the user-facing behavior level.
Content: No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.