Back to skill

Security audit

商品素材库自动混剪|AI-HIVE

Security checks for vulnerabilities and agentic risk

Overview

The skill is a disclosed AI-HIVE video workflow helper that uploads user-chosen media, runs local ffmpeg edits, and may make billable generation calls only through visible commands.

Install only if you are comfortable using AI-HIVE and uploading selected product media to its service. Treat generation as potentially billable, review prompts/routes before running, keep the API key out of logs and repos, and use only media and claims you are authorized to publish.

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
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Behavioral ASTexec() Call, eval() Call, Dynamic Import
  • 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)

Tp4

High
Category
MCP Tool Poisoning
Confidence
97% confidence
Finding

If the skill exposes undeclared capabilities such as generic chat, account or wallet retrieval, broad model inventory access, or general-purpose generation outside its stated montage purpose, users and reviewers cannot accurately assess its access or data-handling behavior. In a skill that already references networked APIs and billable operations, undeclared capability expansion increases the chance of privacy leakage, surprise billing, or misuse of account-scoped functions.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
99% confidence
Finding

If the skill exposes undeclared capabilities such as generic chat, account or wallet retrieval, broad model inventory access, or general-purpose generation outside its stated montage purpose, users and reviewers cannot accurately assess its access or data-handling behavior. In a skill that already references networked APIs and billable operations, undeclared capability expansion increases the chance of privacy leakage, surprise billing, or misuse of account-scoped functions.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
93% confidence
Finding

If the skill exposes undeclared capabilities such as generic chat, account or wallet retrieval, broad model inventory access, or general-purpose generation outside its stated montage purpose, users and reviewers cannot accurately assess its access or data-handling behavior. In a skill that already references networked APIs and billable operations, undeclared capability expansion increases the chance of privacy leakage, surprise billing, or misuse of account-scoped functions.

Content

No source excerpt is available for this finding.

Undeclared Tool Scope

Medium
Category
MCP Least Privilege
Confidence
93% confidence
Finding

The skill advertises executable workflows and includes runnable shell/Python commands that rely on environment variables, filesystem access, network calls, and local tooling, yet it declares no explicit tool scope or permission boundaries. This creates an overbroad and opaque execution surface where an agent or user may run code with capabilities not transparently constrained, increasing risk of unintended file access, secret exposure, or external requests.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The natural-language prompt instructs the skill to produce a "production-ready Chinese workflow," which imposes a specific language choice by default. This is a language/locale policy concern because the file does not indicate any user opt-in or alternative language selection.

Content

No source excerpt is available for this finding.

subprocess module call

Medium
Category
Dangerous Code Execution
Confidence
70% confidence
Finding

subprocess module calls execute external commands. Without careful input validation, this enables command injection.

Content

Scanner excerpt · scripts/edit_video.py (reported line 20)May include surrounding context.

python
def run(args: list[str]) -> None:
    print(" ".join(args))
    subprocess.run(args, check=True)


def main() -> None:

Description-Behavior Mismatch

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The manifest scopes this skill to 商品素材库自动混剪 and related e-commerce montage workflows, but the file-level docstring explicitly presents the code as a '通用 AI 能力调用工具' wrapping text chat, image generation, video generation, model query, media upload, and task polling. That documented and implemented breadth exceeds the manifest’s narrower product-montage intent.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The module docstring and all user-facing CLI guidance are written in Chinese, presenting the skill as a general-purpose tool without stating that it is Chinese-only or offering locale selection. For a broadly applicable skill, this is a natural-language locale policy issue because it effectively forces one language without documented opt-in or justification.

Content

No source excerpt is available for this finding.

Context-Inappropriate Capability

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The manifest describes a workflow for ingesting product media, SKU information, and generating traceable montage outputs via AI-HIVE. A generic chat/text endpoint is implemented and exposed elsewhere in the file, but conversational text generation is not justified as a direct requirement of the stated montage skill purpose.

Content

No source excerpt is available for this finding.

Context-Inappropriate Capability

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

This skill’s manifest is centered on automatic remixing of existing product asset libraries and, when needed, generating supporting materials through AI-HIVE as part of that workflow. A generic image-generation endpoint is implemented as a first-class capability, which is broader than a montage-specific helper and is not exposed as a constrained, purpose-bound substep here.

Content

No source excerpt is available for this finding.

Intent-Code Divergence

Low
Category
Not specified by scanner
Confidence
97% confidence
Finding

The upload command prints that the returned mediaId can be passed to chat/image/video '--image' or '--video' arguments. However, the command handlers treat those arguments as local file paths and call upload_media on them, so supplying a mediaId would fail path validation instead of reusing the uploaded asset.

Content

No source excerpt is available for this finding.