Back to skill

Security audit

Minimax Image

Security checks for vulnerabilities and agentic risk

Overview

This image-generation skill mostly matches its stated purpose, but it can send an AIMLAPI credential to the MiniMax endpoint and logs full prompts in plaintext.

Review before installing. Use only a MiniMax-specific API key, avoid configuring AIMLAPI_API_KEY for this skill unless the provider routing is fixed, and do not include sensitive or confidential details in prompts because prompts are sent to MiniMax and also written to runtime logs. Prefer a version that removes plaintext prompt logging, binds each credential to its own endpoint, and refreshes dependencies from a trusted registry.

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 (3)

T09 · Insecure Skill Coding Practices

Error
Location
index.js:88
Finding

Alternative Provider Credential Is Transmitted to the MiniMax Endpoint

Content
View full analysis
Remediation
View remediation

T09 · Insecure Skill Coding Practices

Note
Location
index.js:108
Finding

User Prompts Are Written to Runtime Logs in Plaintext

Content
View full analysis
Remediation
View remediation

T08 · Insecure Dependencies

Warning
Location
package-lock.json:17
Finding

Locked Dependencies Are Retrieved from a Third-Party Package Mirror

Content
View full analysis
Remediation
View remediation
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • System Prompt LeakageDirect Leakage, Indirect Extraction, Tool-Based Exfiltration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
Findings (16)

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 · index.js (reported line 35)May include surrounding context.

js
// If prompt is already in English, just add quality keywords
    if (/^[a-zA-Z0-9\s,]+$/.test(prompt)) {
        return prompt + ', ' + qualityKeywords;
    }
    
    // Map common Chinese subjects to English

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 · index.js (reported line 91)May include surrounding context.

js
// If prompt is already in English, just add quality keywords
    if (/^[a-zA-Z0-9\s,]+$/.test(prompt)) {
        return prompt + ', ' + qualityKeywords;
    }
    
    // Map common Chinese subjects to English

Known Vulnerable Dependency: axios==1.13.6 — 16 advisory(ies): CVE-2026-44494 (axios Vulnerable to Full Man-in-the-Middle via Prototype Pollution Gadget in `co); CVE-2026-44495 (axios Vulnerable to Credential Theft and Response Hijacking via Prototype Pollut); CVE-2025-62718 (Axios has a NO_PROXY Hostname Normalization Bypass that Leads to SSRF) +13 more

High
Category
Supply Chain
Confidence
96% confidence
Finding

The lockfile pins axios 1.13.6, which the scanner reports as having multiple known advisories including SSRF/proxy-bypass and prototype-pollution-related attack paths. In an image-generation skill that likely makes outbound API requests using user-influenced prompts and possibly remote URLs, a vulnerable HTTP client increases risk of request misrouting, credential leakage, or unsafe redirect/proxy handling.

Content

No source excerpt is available for this finding.

Known Vulnerable Dependency: form-data==4.0.5 — 1 advisory(ies): CVE-2026-12143 (form-data: CRLF injection in form-data via unescaped multipart field names and f)

High
Category
Supply Chain
Confidence
86% confidence
Finding

form-data 4.0.5 is reported as vulnerable to CRLF injection via unescaped multipart field names/filenames. If the skill uploads user-controlled files or constructs multipart requests from user input, this could let an attacker manipulate request boundaries or inject unintended headers/content into outbound requests.

Content

No source excerpt is available for this finding.

Known Vulnerable Dependency: axios==1.13.6 — 16 advisory(ies): CVE-2026-44494 (axios Vulnerable to Full Man-in-the-Middle via Prototype Pollution Gadget in `co); CVE-2026-44495 (axios Vulnerable to Credential Theft and Response Hijacking via Prototype Pollut); CVE-2025-62718 (Axios has a NO_PROXY Hostname Normalization Bypass that Leads to SSRF) +13 more

High
Category
Supply Chain
Confidence
98% confidence
Finding

The package explicitly depends on axios 1.13.6, which the finding indicates is affected by multiple published advisories, including SSRF/proxy-bypass and prototype-pollution-related issues. For an image-generation skill that likely makes outbound API requests and may process user-influenced URLs, headers, or proxy settings, a vulnerable HTTP client increases the risk of request tampering, credential leakage, SSRF, or response hijacking.

Content

No source excerpt is available for this finding.

Undeclared Tool Scope

Medium
Category
MCP Least Privilege
Confidence
89% confidence
Finding

The skill declares access to environment-backed secrets via required API keys but does not define an explicit tool scope or permission boundary. This creates ambiguity about what runtime capabilities the skill needs and can lead to overbroad execution contexts where secrets are accessible without clear user or platform control.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The description uses broad trigger phrases such as general requests to generate pictures or draw, which increases the chance of accidental or overly eager invocation. In an agent setting, broad activation is risky because it can cause unintentional transmission of user prompts to an external service without sufficiently explicit user intent.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The skill sends user prompts to an external API but does not provide a clear user-facing warning in the description or usage notes. Users may supply sensitive, proprietary, or personal information assuming local handling, creating privacy and data-governance risk when that content is transmitted off-platform.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

The skill automatically rewrites prompts into English without user opt-in or justification, which can alter meaning, remove safety qualifiers, or transform sensitive content before external transmission. This is especially risky for creative or safety-sensitive prompts because silent prompt mutation reduces user control and auditability.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
91% confidence
Finding

The skill explicitly transmits user-supplied content to a third-party endpoint, which is a real data egress path. In the context of a general-purpose image tool with broad invocation phrases and no prominent warning, this external transmission meaningfully increases privacy, compliance, and inadvertent disclosure risk.

Content

Scanner excerpt · SKILL.md (reported line 37)May include surrounding context.

md
## API

- Endpoint: `https://api.minimaxi.com/v1/image_generation`
- Model: `image-01`
- Aspect Ratio: 16:9 (default)

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The code is explicitly built around Chinese-only prompt stripping patterns and later returns Chinese-only user-facing messages. This creates a language/locale constraint without any opt-in or documented justification, which matches the policy category for forced language behavior.

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 · index.js (reported line 100)May include surrounding context.

js
return { text: "❌ 未找到 API Key,请设置 MINIMAX_API_KEY 或 AIMLAPI_API_KEY 环境变量。" };
    }

    const API_URL = "https://api.minimaxi.com/v1/image_generation";

    // Get user input
    let userInput = context.args.join(' ').trim();

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

The skill sends user-supplied prompt content to a third-party API without any explicit user-facing disclosure or consent flow. If users include sensitive personal, proprietary, or regulated information in prompts, that data is transmitted externally and may be logged, retained, or processed by the provider.

Content

No source excerpt is available for this finding.

Context-Inappropriate Capability

Low
Category
Not specified by scanner
Confidence
86% confidence
Finding

The manifest frames this skill as generating images and optimizing prompts, but the code also directly accesses process environment variables to retrieve credentials. While needed for implementation, credential access is a capability not stated in the skill purpose and can be context-inappropriate when the manifest does not declare it.

Content

No source excerpt is available for this finding.

Known Vulnerable Dependency: follow-redirects==1.15.11 — 1 advisory(ies): CVE-2026-40895 (follow-redirects leaks Custom Authentication Headers to Cross-Domain Redirect Ta)

Low
Category
Supply Chain
Confidence
88% confidence
Finding

follow-redirects 1.15.11 is flagged for leaking custom authentication headers across cross-domain redirects. If this skill sends authenticated requests to the MiniMax API, an attacker able to trigger or influence redirects could cause API keys or bearer tokens to be forwarded to an unintended host.

Content

No source excerpt is available for this finding.

Unpinned Dependencies

Low
Category
Supply Chain
Confidence
90% confidence
Finding

The dependency is specified with a caret range, which allows automatic installation of newer minor/patch versions without a lock to an exact reviewed release. In a security-sensitive agent skill, this weakens supply-chain integrity and can unexpectedly pull in vulnerable or behavior-changing code during install or rebuilds.

Content

Scanner excerpt · package.json (reported line 13)May include surrounding context.

json
"author": "",
  "license": "ISC",
  "dependencies": {
    "axios": "^1.13.6"
  }
}