Back to skill

Security audit

一人电商运营助手|AI-HIVE

Security checks for vulnerabilities and agentic risk

Overview

The skill mostly matches its AI-HIVE ecommerce content purpose, but it needs review because its scripts can send an API key to a configurable API endpoint and perform media transfers without strong destination or size limits.

Install only if you are comfortable giving the scripts an AI-HIVE API key and uploading selected product media to AI-HIVE/object-storage services. Avoid setting AI_HIVE_BASE_URL or --base-url unless you fully trust the endpoint, review ~/.ai-hive/config.json before use, prefer a limited API key, and monitor generated-task costs and downloaded file sizes.

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:91
Finding

API credentials can be forwarded to an unrestricted endpoint

Content
View full analysis
Remediation
View remediation

T09 · Insecure Skill Coding Practices

Warning
Location
scripts/imagegen.py:266
Finding

Server-controlled upload and download URLs are used without destination or size validation

Content
View full analysis
Remediation
View remediation

T08 · Insecure Dependencies

Note
Location
scripts/imagegen.py:17
Finding

Runtime dependency guidance installs an unpinned package

Content
View full analysis
Remediation
View remediation
Vulnerability Patterns
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • 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
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
Findings (16)

Tp4

High
Category
MCP Tool Poisoning
Confidence
97% confidence
Finding

The skill promises planning outputs based on product, budget, time, platform, and goals, but the behavior described by analysis is largely generic AI-HIVE API access with no actual implementation of the claimed planning logic. A misleading wrapper around generic API operations is dangerous because it can trick users into granting trust and potentially sensitive inputs while the real behavior performs unrelated networked actions and exposes broader account-linked functionality.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
98% confidence
Finding

The skill promises planning outputs based on product, budget, time, platform, and goals, but the behavior described by analysis is largely generic AI-HIVE API access with no actual implementation of the claimed planning logic. A misleading wrapper around generic API operations is dangerous because it can trick users into granting trust and potentially sensitive inputs while the real behavior performs unrelated networked actions and exposes broader account-linked functionality.

Content

No source excerpt is available for this finding.

Undeclared Tool Scope

Medium
Category
MCP Least Privilege
Confidence
84% confidence
Finding

The skill advertises executable scripts and networked API workflows but does not declare any tool scope or allowed-tools boundary. That makes the operational capability opaque to the host and reviewers, increasing the chance of over-privileged execution, unsafe file/network access, or accidental secret exposure when the skill is invoked.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
89% confidence
Finding

Broad trigger phrases like personal seller, low-cost operations, or ecommerce assistant can cause the skill to activate in routine conversations that do not require its higher-risk media-generation and networked behavior. Over-triggering increases the likelihood of accidental invocation, unnecessary exposure of user business data, and unintended progression into billable or more privileged actions.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
97% confidence
Finding

The natural-language instruction says to turn requests into a "production-ready Chinese workflow," which imposes a specific language on all outputs. This is a locale/language policy concern because the file does not offer the user any language choice or opt-in mechanism.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

Medium
Category
Not specified by scanner
Confidence
97% confidence
Finding

The file presents itself as a focused solo-ecommerce/image skill, but most of the implementation is a general-purpose AI Hive client supporting chat, model enumeration, uploads, and video generation. This capability mismatch increases the attack surface and can enable use beyond the user's expected consent boundary, especially in agent environments that rely on metadata to decide what a skill may do.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

Natural-language strings throughout the skill, including the module description, CLI help, prompts, and status messages, are fixed to Chinese. Because the file does not offer locale selection or document a justified region-specific constraint, this can violate language/locale policy requiring user choice or explicit opt-in.

Content

No source excerpt is available for this finding.

Context-Inappropriate Capability

Medium
Category
Not specified by scanner
Confidence
88% confidence
Finding

This file implements a full chat_text API path and CLI command for arbitrary text or multimodal conversations. The wrapped skill configuration and manifest focus on a solo-ecommerce operational assistant with AI-Hive used for asset generation, especially image-generation workflow details, so unrestricted generic chat is broader than the declared purpose of this specific skill file.

Content

No source excerpt is available for this finding.

Context-Inappropriate Capability

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The concrete skill wrapper in this file is titled and wired as an image skill, with SKILL_CONFIG example text describing commercial image creation. Yet the module also contains a complete video-generation client and CLI path, which goes beyond the declared image-generation role of this skill and broadens capability into a separate media modality.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The skill uploads local media to remote services and downloads generated outputs from returned URLs without an explicit user-facing privacy notice or transmission confirmation. In an agent skill context handling business creatives and reference assets, this can cause unintended disclosure of sensitive images, product materials, or other proprietary content to third-party infrastructure.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

The file advertises and implements a generic AI Hive OpenAPI client with chat, image, video, model listing, media upload, and task polling, which is materially broader than the manifest’s stated one-person ecommerce video assistant use case. In a skill ecosystem, this scope expansion increases the attack surface and enables capability abuse outside the user’s expected consent boundary, even if the code is not overtly malicious.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
90% confidence
Finding

This code file contains natural-language strings that present the skill entirely in Chinese, including the top-level description and argument parser description. Under the policy criteria, forcing a specific language without user opt-in can be a locale/language policy violation unless the regional constraint is explicitly justified, which is not stated here.

Content

No source excerpt is available for this finding.

Context-Inappropriate Capability

Low
Category
Not specified by scanner
Confidence
86% confidence
Finding

The manifest emphasizes helping sellers generate low-cost operational and marketing assets through AI-Hive APIs. Exposing a dedicated user-info endpoint for querying account details and wallet balance is an account-management capability that is not necessary to fulfill that stated assistant purpose.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
97% confidence
Finding

This code presents its top-level skill description entirely in Chinese, and the CLI throughout the file is also Chinese-only. That creates a language/locale constraint without any opt-in or documented justification, which matches the policy-violation category for forced language behavior.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

Low
Category
Not specified by scanner
Confidence
92% confidence
Finding

The code exposes a user-info endpoint that can reveal account details and wallet/balance information not described in the skill manifest. While this is not a direct code-execution flaw, it violates least privilege and creates unnecessary access to potentially sensitive account metadata if the skill is invoked or repurposed unexpectedly.

Content

No source excerpt is available for this finding.

Context-Inappropriate Capability

Low
Category
Not specified by scanner
Confidence
86% confidence
Finding

The init flow opens a browser to an external site and guides the user through account login and API key creation. Although credential setup may be convenient for a CLI utility, it is not part of the manifest’s stated assistant behavior around ecommerce planning and AI-Hive generation tasks.

Content

No source excerpt is available for this finding.