Back to skill

Security audit

AI大模型专家|电商视频

Security checks for vulnerabilities and agentic risk

Overview

The skill’s AI-HIVE video workflow is mostly coherent, but it needs Review because it can upload user media/run remote jobs and its configurable API endpoint can expose the AI-HIVE API key to an arbitrary host.

Install only if you are comfortable sending prompts and selected media to AI-HIVE and potentially incurring service charges. Use a dedicated limited API key, confirm the configured base_url is the legitimate AI-HIVE endpoint before running, avoid sensitive customer or unreleased product assets unless approved, and review task costs before bulk generation.

Vulnerability Patterns
  • 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
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
Findings (2)

T09 · Insecure Skill Coding Practices

Error
Location
scripts/videogen.py:67
Finding

API Key Disclosure Through Unrestricted API Base URL Override

Content
View full analysis
Remediation
View remediation

T09 · Insecure Skill Coding Practices

Warning
Location
scripts/videogen.py:643
Finding

Non-Atomic API Key File Creation and Symlink Following

Content
View full analysis
Remediation
View remediation
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Findings (11)

Undeclared Tool Scope

Medium
Category
MCP Least Privilege
Confidence
91% confidence
Finding

The skill describes capabilities that can read environment/config, write files, access the network, and invoke shell commands, but it does not declare any explicit tool scope or permission boundaries. In an agent environment, this increases the chance that the skill is invoked with broader authority than users expect, enabling unintended file access, secret exposure, remote uploads, or command execution through normal-looking video-generation flows.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The invocation description is broad enough to match many generic user requests about video creation, which can cause the skill to be selected in situations where the user did not intend to use this specific remote service. Because the skill also performs uploads, task submission, and downloads, overbroad triggering can lead to unanticipated data transmission and paid actions.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
88% confidence
Finding

The metadata explicitly sets the skill language to zh-CN, while the document also includes English trigger text and does not explain that the skill is limited to Chinese-only operation. This appears to impose a locale preference without user opt-in or a clearly documented regional compliance reason.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The skill states that it automatically uploads media, submits tasks, polls status, and downloads results, but it does not prominently warn that user-provided images, video, audio, prompts, and related metadata will be transmitted to a third-party remote platform. This can cause users to disclose sensitive or proprietary assets without informed consent, especially in commercial ecommerce workflows where product media, unreleased campaigns, or customer-related content may be confidential.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
89% confidence
Finding

The search coverage lists many generic platforms, models, and content-creation terms without constraints, increasing the probability of accidental or opportunistic activation. In this context, mistaken routing is security-relevant because activation may trigger remote media upload, account use, billing, and persistence of task metadata.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The skill enables implicit invocation with no trigger constraints, exclusions, or narrowing conditions, which can cause the agent to invoke this skill automatically in loosely related conversations. Because this skill submits media-generation tasks, uploads user-provided assets, polls remote jobs, and downloads results through AI-HIVE, unintended invocation can expose user data to an external service and trigger unwanted actions or costs.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The manifest frames this skill as focused on 电商视频 / video-production workflows, but the file implements a broader generic client with text chat, image generation, model enumeration, user-info retrieval, and generic media upload capabilities. While network access and file transfer are expected for the stated purpose, these extra non-video operations materially broaden the behavior beyond the skill's declared e-commerce-video scope.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The module docstring and user-facing CLI messaging are written entirely in Chinese, presenting the skill as a general-purpose tool rather than a region-specific one. Because the file does not offer an English or user-selected locale option, it imposes a language choice on all users, which matches the language/locale policy violation criterion.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
85% confidence
Finding

The visible skill metadata is written in Chinese, but the file does not indicate that language selection is user-configurable or that the skill is intentionally limited to a Chinese-speaking context. This can violate language or locale policy where user opt-in or clear regional scoping is required.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
95% confidence
Finding

The example API key placeholder uses Chinese text ("请替换为你的APIKey"), which imposes a specific language in user-facing configuration content. The file does not offer a language choice or explain that the skill is intentionally region- or locale-specific.

Content

No source excerpt is available for this finding.

Intent-Code Divergence

Low
Category
Not specified by scanner
Confidence
91% confidence
Finding

The top-level docstring says this is a '通用 AI 能力调用工具' that wraps text chat, image generation, video generation, model queries, media upload, and polling. That directly conflicts with the specialized skill manifest and skill-specific configuration at the bottom of the file, which present the artifact as an e-commerce video skill rather than a general-purpose AI client.

Content

No source excerpt is available for this finding.