Back to skill

Security audit

用户晒单素材转广告|AI-HIVE

Security checks for vulnerabilities and agentic risk

Overview

The skill matches its advertised AI-HIVE ad-video workflow, but it handles API keys, uploads media, and can start billable generation with weaker safeguards than its documentation implies.

Review this skill before installing. Only use it with an AI-HIVE API key you are willing to use from this machine, keep the base URL at the official endpoint, check generated commands before running them, use separate output paths for edits, and confirm cost/model parameters before any generation task is submitted.

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 (3)

T09 · Insecure Skill Coding Practices

Error
Location
scripts/videogen.py:91
Finding

API Key Can Be Redirected to an Arbitrary Server

Content
View full analysis
Remediation
View remediation

T09 · Insecure Skill Coding Practices

Warning
Location
scripts/edit_video.py:63
Finding

FFmpeg Operations Unconditionally Overwrite Existing Files

Content
View full analysis
Remediation
View remediation

T09 · Insecure Skill Coding Practices

Warning
Location
scripts/videogen.py:503
Finding

Potentially Billable Generation Tasks Are Submitted Without Confirmation

Content
View full analysis
{forwarded.model}") cmd_video(client, forwarded) ``` ### Technical Analysis The script retrieves a pricing snapshot before generation, but it does not display that snapshot, summarize the selected model and parameters, or ...[truncated 1500 chars]
Remediation
View remediation
Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • 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 (12)

Tp4

High
Category
MCP Tool Poisoning
Confidence
97% confidence
Finding

This finding expands the mismatch to include undeclared generic capabilities such as text/image generation, account or wallet access, model enumeration, and interactive API-key setup. Undeclared capabilities increase security risk because operators may approve the skill for a narrow remix use case while it can access broader functions with privacy, billing, or data-handling consequences.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
96% confidence
Finding

This finding expands the mismatch to include undeclared generic capabilities such as text/image generation, account or wallet access, model enumeration, and interactive API-key setup. Undeclared capabilities increase security risk because operators may approve the skill for a narrow remix use case while it can access broader functions with privacy, billing, or data-handling consequences.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
91% confidence
Finding

This finding expands the mismatch to include undeclared generic capabilities such as text/image generation, account or wallet access, model enumeration, and interactive API-key setup. Undeclared capabilities increase security risk because operators may approve the skill for a narrow remix use case while it can access broader functions with privacy, billing, or data-handling consequences.

Content

No source excerpt is available for this finding.

Undeclared Tool Scope

Medium
Category
MCP Least Privilege
Confidence
94% confidence
Finding

The skill advertises and demonstrates code execution patterns involving environment variables, local file operations, networking, and shell commands, but it does not declare any explicit tool scope or allowed-tools boundary. In an agent environment, missing permission declarations can lead to overbroad runtime access, making accidental or unauthorized network calls, file modification, or secret exposure more likely.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
85% confidence
Finding

The skill description and operating instructions are entirely framed in Chinese and position the skill for Chinese-language use, but do not state that language is optional or user-selectable. This can amount to a locale/language policy issue because the skill appears to impose a specific language without explicit user opt-in.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The trigger phrases include broad everyday terms like UGC素材、客户视频、电商广告 that can match many benign requests outside the intended high-friction workflow. Overbroad invocation can cause the wrong skill to activate, leading to unintended file/network actions, user confusion, or accidental routing into a potentially billable external service.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The default prompt and implicit invocation setting make the skill easy to trigger for loosely related requests, which can cause the agent to inject this workflow when the user did not explicitly ask for it. In a skill that produces runnable AI-HIVE commands and ad-production workflows, over-broad activation increases the risk of unwanted tool use, scope creep, and accidental generation of marketing or media-processing actions from ambiguous input.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
97% confidence
Finding

Forcing Chinese output in the default prompt overrides user preference and can cause misunderstanding of generated workflow steps, safety guidance, or runnable commands. While this is not directly a code-execution flaw, it can degrade informed user review and increase the chance that users run commands or approve deliverables they do not fully understand.

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
96% confidence
Finding

The file is presented as a narrowly scoped skill for turning authorized customer-post/UGC materials into ad videos, but it actually exposes a broad general-purpose AI client with text chat, image generation, generic video generation, model enumeration, arbitrary media upload, and task polling. This scope expansion weakens least-privilege boundaries and makes it easier for downstream agents or users to invoke capabilities outside the declared business purpose, increasing the chance of policy bypass, misuse, or unintended data handling.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

This code file contains user-facing natural-language descriptions and prompts entirely in Chinese, presenting the tool as a general-purpose skill rather than a clearly region-specific one. Because it does not offer a language/locale option or document a justified locale restriction, it may violate the policy against forcing a specific language without user opt-in.

Content

No source excerpt is available for this finding.

Context-Inappropriate Capability

Low
Category
Not specified by scanner
Confidence
89% confidence
Finding

The skill includes a user-info capability that retrieves account information and wallet balance even though that functionality is unrelated to ad-production. While not an exploit by itself, exposing account metadata unnecessarily increases information disclosure and broadens what an invoking agent can inspect beyond the declared task.

Content

No source excerpt is available for this finding.