Back to skill

Security audit

视觉解压循环广告|AI-HIVE

Security checks for vulnerabilities and agentic risk

Overview

This appears to be a real AI-HIVE video-generation skill, but it needs Review because its API key can be sent to a configurable server and invocation scope is broad.

Install only if you trust the publisher and the AI-HIVE endpoint. Avoid using custom --base-url or AI_HIVE_BASE_URL values, review ~/.ai-hive/config.json before authenticated use, and confirm any generation command because API calls may upload media and incur charges.

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 (2)

T09 · Insecure Skill Coding Practices

Warning
Location
scripts/videogen.py:92
Finding

Bearer Credential Disclosure Through a User-Controlled API Base URL

Content
View full analysis
Remediation
View remediation

T08 · Insecure Dependencies

Note
Location
scripts/videogen.py:9
Finding

Unpinned Third-Party Dependency Installation Guidance

Content
View full analysis
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 (10)

Tp4

High
Category
MCP Tool Poisoning
Confidence
98% confidence
Finding

The documented behavior promises specialized ad-loop design, continuity checks, originality safeguards, and AI-HIVE integration, but the detectable implementation appears to provide only generic editing examples. This mismatch is dangerous because operators may trust the skill to enforce business, safety, or authorization checks that are not actually implemented, leading to misuse, unexpected billing, weak auditability, or policy bypass through misplaced reliance.

Content

No source excerpt is available for this finding.

Undeclared Tool Scope

Medium
Category
MCP Least Privilege
Confidence
92% confidence
Finding

The skill advertises executable workflows involving environment variables, local file access, shell commands, and network calls, but it declares no explicit tool scope or permission boundaries. In an agent environment, this can lead to overbroad capability exposure, making it easier for the skill to access secrets, modify files, or reach external services without clear least-privilege constraints.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
87% confidence
Finding

The listed activation trigger 短视频创意 is a broad, common phrase that could match many ordinary requests unrelated to this specific looping ad skill. The surrounding section provides examples, but it does not clearly constrain how this trigger should be distinguished from general short-video ideation requests.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
98% confidence
Finding

The default prompt hard-codes Chinese output and does not preserve user language preference or require opt-in. This is primarily a trust, usability, and prompt-governance issue: it can override user expectations, obscure safety-critical instructions if the user expects another language, and make unintended autonomous behavior less transparent when combined with implicit invocation.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The manifest allows implicit invocation without any visible scoping constraints, exclusions, or tighter activation criteria. In a skill that can generate production-ready workflows and runnable AI-HIVE commands, this increases the chance the skill is invoked in contexts the user did not clearly intend, causing unwanted API actions, content generation, or policy-bypassing assistance to be surfaced automatically.

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 embeds a broad, reusable AI Hive client with capabilities for text chat, image generation, model enumeration, uploads, and account queries, while the advertised skill is specifically for looped satisfying-ad video generation. This scope mismatch increases attack surface and enables capability creep: a caller who can invoke the script may access unrelated functions and external API actions not necessary for the declared purpose.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
84% confidence
Finding

The file hard-codes Chinese-language title/branding in the module docstring, argument parser description, and generated JSON skill field. This imposes a specific language presentation without any visible opt-in or alternative locale selection, which matches the language/locale policy concern.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
95% confidence
Finding

The module docstring presents the skill description entirely in Chinese, and later user-facing help text is also Chinese-only. Under the policy, forcing a specific language without user opt-in is a natural-language policy violation when no alternative locale or language choice is offered.

Content

No source excerpt is available for this finding.

Context-Inappropriate Capability

Low
Category
Not specified by scanner
Confidence
91% confidence
Finding

Exposing a user-info endpoint that includes account and wallet/balance data is unnecessary for producing looped advertising videos and creates an avoidable data exposure path. Even if limited to the current account, it discloses sensitive operational metadata that could aid reconnaissance, billing abuse awareness, or privacy violations if the skill is invoked in a broader agent environment.

Content

No source excerpt is available for this finding.