Back to skill

Security audit

AI虚拟网红孵化器|AI-HIVE

Security checks for vulnerabilities and agentic risk

Overview

The skill matches its AI-HIVE media-generation purpose, but it needs Review because it handles API keys and user media with under-scoped endpoint and download safeguards.

Install only if you trust AI-HIVE and the publisher, are comfortable storing an API key locally, and understand that selected media may be uploaded to third-party services and generation may incur charges. Avoid using custom base URLs or untrusted environments, use a dedicated low-privilege API key if possible, and review outputs/download locations before using it with sensitive brand, likeness, customer, or proprietary media.

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 (3)

T09 · Insecure Skill Coding Practices

Error
Location
scripts/imagegen.py:92
Finding

Bearer Credential Disclosure Through an Unrestricted API Base URL

Content
View full analysis
Remediation
View remediation

T09 · Insecure Skill Coding Practices

Warning
Location
scripts/imagegen.py:266
Finding

Unvalidated Server-Controlled Upload and Download Destinations

Content
View full analysis
Remediation
View remediation

T08 · Insecure Dependencies

Note
Location
scripts/imagegen.py:7
Finding

Unpinned Third-Party Runtime Dependency

Content
View full analysis
Remediation
View remediation
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • 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
Findings (26)

Tp4

High
Category
MCP Tool Poisoning
Confidence
93% confidence
Finding

This mismatch includes generic text/image generation, media upload, browser opening, and local API key storage/setup, which is substantially broader than a narrowly described virtual-influencer incubator. In this context, the combination of uploads, local configuration persistence, and broader model access raises the risk of unintended data exfiltration, credential mishandling, and user approval based on misleading descriptions.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
99% confidence
Finding

This mismatch includes generic text/image generation, media upload, browser opening, and local API key storage/setup, which is substantially broader than a narrowly described virtual-influencer incubator. In this context, the combination of uploads, local configuration persistence, and broader model access raises the risk of unintended data exfiltration, credential mishandling, and user approval based on misleading descriptions.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
95% confidence
Finding

This mismatch includes generic text/image generation, media upload, browser opening, and local API key storage/setup, which is substantially broader than a narrowly described virtual-influencer incubator. In this context, the combination of uploads, local configuration persistence, and broader model access raises the risk of unintended data exfiltration, credential mishandling, and user approval based on misleading descriptions.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
95% confidence
Finding

This mismatch includes generic text/image generation, media upload, browser opening, and local API key storage/setup, which is substantially broader than a narrowly described virtual-influencer incubator. In this context, the combination of uploads, local configuration persistence, and broader model access raises the risk of unintended data exfiltration, credential mishandling, and user approval based on misleading descriptions.

Content

No source excerpt is available for this finding.

Ae1

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 81)May include surrounding context.

md
python3 scripts/imagegen.py init --skill-name ai-virtual-influencer-incubator-ai-hive

Ae1

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 83)May include surrounding context.

md
python3 scripts/imagegen.py init --skill-name ai-virtual-influencer-incubator-ai-hive

Ae1

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 93)May include surrounding context.

md
python3 scripts/videogen.py init --skill-name ai-virtual-influencer-incubator-ai-hive

Ae1

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 95)May include surrounding context.

md
python3 scripts/videogen.py init --skill-name ai-virtual-influencer-incubator-ai-hive

Undeclared Tool Scope

Medium
Category
MCP Least Privilege
Confidence
90% confidence
Finding

The skill advertises executable commands and code paths that can read environment variables, access local files, invoke shell tools, and make network requests, but it does not declare any explicit tool scope or permissions boundary. That increases the risk of over-broad execution in hosts that rely on manifest-level restrictions, especially because the documented workflow includes API key handling, uploads, downloads, and ffmpeg execution.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

The phrase production-ready Chinese workflow imposes a specific language on the skill's output. Under the language/locale policy, this is a violation unless the skill offers a language choice, obtains user opt-in, or clearly documents a justified locale restriction.

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 visible trigger constraints, so it may activate in response to broad or ambiguous user requests. In a high-capability content-generation skill that can produce workflows and runnable AI-HIVE commands, over-broad activation increases the chance of unintended execution paths, surprise tool use, or the assistant steering users into this skill without clear consent.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

The script’s natural-language strings and defaults are hard-coded in Chinese, including the skill name, search aliases, and default platform value, with no option for the user to choose language or locale. This can violate language/locale policy because the skill effectively imposes one language regardless of user preference.

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
93% confidence
Finding

The file presents itself as a domain-specific virtual-influencer incubator skill, but actually embeds a broad AI Hive client exposing unrelated capabilities such as account inspection, text chat, model enumeration, media upload, and video generation. This capability mismatch violates least privilege and expands the skill surface beyond what users and reviewers would reasonably expect, making abuse or policy bypass easier if the skill is trusted for a narrower purpose.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
97% confidence
Finding

The module docstring and user-facing instructions are entirely in Chinese, and the CLI prompts/messages throughout the skill assume Chinese-language interaction. This constitutes a language/locale policy concern because the skill does not offer opt-in language selection or document that it is intentionally limited to a Chinese-speaking or region-specific context.

Content

No source excerpt is available for this finding.

Context-Inappropriate Capability

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

The generic text-chat functionality is unrelated to the declared image-generation skill surface and materially increases the tool's power and ambiguity. A broader capability set allows the skill to be repurposed for unintended workflows, weakens user consent boundaries, and makes it harder to apply policy controls tied to the manifest's stated purpose.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The file presents itself as a narrowly scoped virtual-influencer incubator skill, but embeds a much broader generic AI client with chat, model enumeration, account inspection, and arbitrary media upload primitives. That scope expansion increases the attack surface and enables callers to use the skill for unrelated capabilities that may expose data or bypass the intended product boundaries.

Content

No source excerpt is available for this finding.

Context-Inappropriate Capability

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The skill can query user account information and wallet balance even though that capability is not needed for video generation. In a skill context, unnecessary access to billing/account metadata violates least privilege and can leak sensitive operational information to downstream users or prompts that should only trigger content generation.

Content

No source excerpt is available for this finding.

Context-Inappropriate Capability

Medium
Category
Not specified by scanner
Confidence
90% confidence
Finding

A standalone generic text chat endpoint is exposed despite the skill being described as a virtual-influencer incubator/video generation tool. This allows repurposing the skill as a general LLM proxy, undermining scoping assumptions, increasing misuse opportunities, and potentially exposing API-funded capabilities outside the intended workflow.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
89% confidence
Finding

The upload flow sends local media files to remote object storage and the AI Hive service, but the command interface does not prominently warn users that their files will leave the local machine. In this skill context, users may provide sensitive brand assets, likenesses, or personal media, so lack of explicit transmission notice increases the chance of unintentional data disclosure.

Content

No source excerpt is available for this finding.

Intent-Code Divergence

Low
Category
Not specified by scanner
Confidence
84% confidence
Finding

The docstring says the script creates a deterministic production brief, which implies a more complete finalized deliverable. In practice, the code emits a basic JSON object with status set to "draft" and a limited subset of planning metadata, so the documentation overstates what is produced.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

Low
Category
Not specified by scanner
Confidence
90% confidence
Finding

The manifest describes a skill that accepts rich incubator inputs and can deliver role positioning, character bible, visual anchors, content matrices, media generation workflow, and AI-HIVE API interactions. This script only builds a small local JSON draft with fixed fields and writes it to disk, without any API configuration, asset upload, routing, polling, download, or broader incubator outputs.

Content

No source excerpt is available for this finding.

Intent-Code Divergence

Low
Category
Not specified by scanner
Confidence
88% confidence
Finding

The manifest presents a narrowly purposed skill for AI virtual influencer incubation with explicit domain boundaries and misuse constraints. By contrast, the top-level docstring describes the code as a general-purpose AI Hive tool wrapping chat, image, video, model query, upload, and polling, actively signaling a broader intent than the manifest.

Content

No source excerpt is available for this finding.

Context-Inappropriate Capability

Low
Category
Not specified by scanner
Confidence
90% confidence
Finding

The user-info capability exposes account and wallet/balance inspection that is not necessary for an image-centric virtual-influencer incubation skill. Even if authenticated by the user's own API key, unnecessary access to billing/account metadata increases data exposure and broadens what a compromised or misused skill can query.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
90% confidence
Finding

The module docstring and user-facing CLI guidance are written entirely in Chinese, and the skill does not indicate any language choice or opt-in for other locales. Under the policy rule, forcing a specific language without user choice is a natural-language policy concern.

Content

No source excerpt is available for this finding.