Back to skill

Security audit

虚拟主播商品播报视频|AI-HIVE

Security checks for vulnerabilities and agentic risk

Overview

The skill’s video-generation purpose is mostly coherent, but its API-key routing and download handling create real account and file-safety risks that should be reviewed before installation.

Install only if you trust AI-HIVE and need this workflow. Avoid setting AI_HIVE_BASE_URL or --base-url unless you fully trust the endpoint, keep the API key scoped and revocable, review commands before paid generation or media upload, and download outputs into a low-risk directory.

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:91
Finding

Bearer API Key Can Be Exfiltrated Through an Unrestricted Custom Base URL

Content
View full analysis
Remediation
View remediation

T09 · Insecure Skill Coding Practices

Warning
Location
scripts/videogen.py:399
Finding

API-Controlled Task Metadata Can Escape the Download Directory

Content
View full analysis
Remediation
View remediation
Vulnerability Patterns
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • 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
Findings (13)

Tp4

High
Category
MCP Tool Poisoning
Confidence
95% confidence
Finding

The description promises a substantial workflow for producing virtual-host product broadcast videos, including content generation and external API-based media job orchestration. The actual code only parses command-line arguments and writes a local JSON file labeled as a draft production blueprint. While a blueprint could be a supporting component of such a system, this chunk’s actual behavior is materially narrower than the declared purpose and lacks the key advertised capabilities. Therefore, the description does not accurately represent what this code chunk actually does.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
99% confidence
Finding

This is a clear mismatch. The declared description presents a specialized workflow for generating virtual-host product-promotion videos using external AI APIs and business inputs. The supplied code instead implements a narrow, generic media-editing CLI around ffmpeg/ffprobe. Its operations are limited to probe, trim, aspect conversion, mute, loudness normalization, and concatenation. These are generic support edits and do not realize the declared core functionality. There are no network calls, no AI generation, no content creation pipeline, no e-commerce semantics, and no API orchestration mentioned in the description.

Content

No source excerpt is available for this finding.

Undeclared Tool Scope

Medium
Category
MCP Least Privilege
Confidence
91% confidence
Finding

The skill documents executable commands that use environment variables, local files, networking, and shell-based tooling, but it does not declare any explicit tool scope such as permissions or allowed-tools. That creates an authorization ambiguity where an agent or runtime may grant broader capabilities than reviewers or users expect, increasing the chance of unintended network calls, file access, or secret exposure during execution.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
86% confidence
Finding

The skill instructions and invocation guidance are entirely written in Chinese and repeatedly frame usage around Chinese-language triggers, but do not explicitly state that the skill is Chinese-only or offer a language/locale choice. Under the policy, a skill that effectively forces a specific language without user opt-in is a natural-language policy violation.

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 instruction in the default prompt requires a "production-ready Chinese workflow," which imposes a specific language choice by default. This is a locale/language policy concern because the file does not indicate user choice, opt-in, or a justified region-specific constraint.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
90% confidence
Finding

Enabling implicit invocation without narrowly defined triggers increases the chance that the skill will activate on loosely related requests and steer the conversation into generating runnable AI-HIVE commands and production workflows. In a skill that can upload reference media, query pricing, poll async jobs, and download outputs, accidental activation can lead to unintended third-party API use, unexpected cost-incurring actions, or processing of sensitive media without clear user intent.

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
97% confidence
Finding

The file is presented as a specialized virtual-host product-video skill, but the implementation is a general-purpose AI Hive client that exposes unrelated capabilities. This scope expansion increases the reachable attack surface and enables users or downstream agents to invoke unintended functions outside the reviewed skill contract, weakening policy and permission boundaries.

Content

No source excerpt is available for this finding.

Intent-Code Divergence

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The module self-documents as a generic AI capability tool, contradicting the specialized identity of the skill. That mismatch is dangerous because reviewers and users may trust the manifest's narrow purpose while the code actually supports a much broader set of actions, undermining transparency and security review assumptions.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

This code file contains user-facing natural-language descriptions entirely in Chinese, presenting the tool as a general-purpose skill rather than a region-specific one. Because there is no opt-in or alternative locale offered in the visible instructions and prompts, it creates a language/locale policy concern under the natural-language policy rule.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

Standalone text-chat and image-generation commands materially exceed the declared purpose of a virtual-host product-video skill. In skill ecosystems, this kind of hidden multifunctionality can be abused to bypass capability scoping, invoke unreviewed model classes, or route sensitive prompts and media through APIs that users did not intend to authorize.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
91% confidence
Finding

This Python file contains user-facing natural-language strings entirely in Chinese, including the module docstring, argument parser description, skill name, and search aliases. Under the policy rule for language or locale constraints, this is a violation because the skill does not provide any user opt-in or alternative language choice.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

Low
Category
Not specified by scanner
Confidence
88% confidence
Finding

The skill exposes a user-info/account inspection endpoint that is not part of the declared video-generation purpose. While not directly an exploit by itself, it allows unnecessary access to account metadata and balances, expanding information exposure beyond least privilege for this skill.

Content

No source excerpt is available for this finding.