Back to skill

Security audit

chrome-extension-source

Security checks for vulnerabilities and agentic risk

Overview

This is a straightforward instruction-only skill for downloading and unpacking Chrome extension CRX files, with expected network and local file handling but no hidden execution, persistence, or credential harvesting.

Install only if you need a Chinese-language recipe for CRX extraction. Prefer the official Google download endpoint, use an isolated working directory, treat downloaded and unpacked extension files as untrusted, avoid third-party mirrors unless necessary, and do not share local proxy settings or downloaded proprietary code unless you are authorized to do so.

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
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (5)

External Script Fetching

High
Category
Supply Chain
Confidence
90% confidence
Finding

The skill fetches a remote CRX package and then instructs the user to strip the header and unpack its ZIP payload without performing robust trust validation on the downloaded content. Although the content is not executed directly in the steps shown, this workflow normalizes retrieval of untrusted code artifacts and later suggests local serving or validation, which can lead to analysis of tampered or mirror-substituted content and possible downstream compromise if users inspect or run it unsafely.

Content

Scanner excerpt · skill.md (reported line 28)May include surrounding context.

2a. 直连(官方端点,首选)

bash
curl -sL -o ext.crx \
  "https://clients2.google.com/service/update2/crx?response=redirect&os=win&arch=x86-64&os_arch=x86-64&nacl_arch=x86-64&prod=chromiumcrx&prodchannel=unknown&prodversion=131.0.0.0&acceptformat=crx2,crx3&x=id%3D${EXT_ID}%26uc"

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
84% confidence
Finding

The skill instructs users to download full extension packages from Google and a third-party mirror, then unpack them locally, but it omits strong warnings about proprietary code handling and the privacy/integrity risks of mirror use. In particular, the mirror path could expose the target extension being analyzed to a third party and may return substituted content, making the workflow riskier than presented.

Content

No source excerpt is available for this finding.

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 28)May include surrounding context.

2a. 直连(官方端点,首选)

bash
curl -sL -o ext.crx \
  "https://clients2.google.com/service/update2/crx?response=redirect&os=win&arch=x86-64&os_arch=x86-64&nacl_arch=x86-64&prod=chromiumcrx&prodchannel=unknown&prodversion=131.0.0.0&acceptformat=crx2,crx3&x=id%3D${EXT_ID}%26uc"

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
94% confidence
Finding

All user-facing instructional text is in Chinese, and the document does not provide an alternative language option or state that the skill is intended only for Chinese-speaking users. This can violate language-choice policy where skills should not force a specific language without user opt-in.

Content

No source excerpt is available for this finding.

Missing User Warnings

Low
Category
Not specified by scanner
Confidence
80% confidence
Finding

These lines direct the user to inspect HTTP(S)_PROXY values and a registry key that may contain internal proxy hostnames or credentials, which can be sensitive operational information. The document provides procedural steps but does not include any caution about exposing or sharing that data.

Content

No source excerpt is available for this finding.