Back to skill

Security audit

AI大模型专家|fal.ai 替代方案|AI-HIVE

Security checks for vulnerabilities and agentic risk

Overview

The skill mostly matches an AI-HIVE migration/media workflow, but it handles API keys and remote uploads/downloads with under-scoped safeguards that users should review before installing.

Review this before installing if you will use real AI-HIVE credentials or sensitive media. Do not set --base-url or AI_HIVE_BASE_URL to an untrusted host, avoid uploading confidential assets unless you trust the service and storage path, and use a limited-scope API key with spending controls where possible.

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:94
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:274
Finding

Unvalidated server-provided URLs permit internal requests and unrestricted downloads

Content
View full analysis
Remediation
View remediation

T08 · Insecure Dependencies

Note
Location
scripts/imagegen.py:7
Finding

Unpinned third-party dependency installation

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
  • Behavioral ASTexec() Call, eval() Call, Dynamic Import
Findings (21)

Tp4

High
Category
MCP Tool Poisoning
Confidence
95% confidence
Finding

The reported behavior extends beyond migration consulting into general AI-HIVE account and content operations, including wallet/user-info access, filesystem interaction, browser/config access, and remote object storage usage. Such undisclosed privileged actions can lead to data exfiltration, accidental spending, and unauthorized manipulation because the skill's stated purpose does not prepare users for those risks.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
99% confidence
Finding

The reported behavior extends beyond migration consulting into general AI-HIVE account and content operations, including wallet/user-info access, filesystem interaction, browser/config access, and remote object storage usage. Such undisclosed privileged actions can lead to data exfiltration, accidental spending, and unauthorized manipulation because the skill's stated purpose does not prepare users for those risks.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
95% confidence
Finding

The reported behavior extends beyond migration consulting into general AI-HIVE account and content operations, including wallet/user-info access, filesystem interaction, browser/config access, and remote object storage usage. Such undisclosed privileged actions can lead to data exfiltration, accidental spending, and unauthorized manipulation because the skill's stated purpose does not prepare users for those risks.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
95% confidence
Finding

The reported behavior extends beyond migration consulting into general AI-HIVE account and content operations, including wallet/user-info access, filesystem interaction, browser/config access, and remote object storage usage. Such undisclosed privileged actions can lead to data exfiltration, accidental spending, and unauthorized manipulation because the skill's stated purpose does not prepare users for those risks.

Content

No source excerpt is available for this finding.

Undeclared Tool Scope

Medium
Category
MCP Least Privilege
Confidence
86% confidence
Finding

The skill advertises executable scripts, environment-variable secrets, filesystem paths, shell commands, and networked API usage, but does not declare any explicit tool scope or permission boundaries. This creates an over-privileged and under-specified execution surface where an agent may use sensitive capabilities without clear user-reviewable constraints.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The trigger phrases include broad generic terms such as model API, API aggregation, image API, and video API, which are likely to match many unrelated user requests. Over-broad invocation is dangerous here because the skill appears to have execution, network, and file-handling capabilities, so accidental activation could expose users to unintended data access or billable 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 default prompt hard-requires Chinese output and runnable Chinese examples without checking user language preference. This can override user expectations, reduce clarity for non-Chinese-speaking users, and in safety-sensitive migration guidance increase the risk of misunderstanding technical steps, acceptance criteria, or code behavior.

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, exclusions, or user-confirmation guardrails. That increases the chance the agent auto-selects this migration/advisory skill in loosely related conversations, causing unsolicited steering toward a third-party platform and potentially shaping architectural or purchasing decisions without clear user intent.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The manifest describes a skill that helps evaluate AI-HIVE as an alternative through compatibility auditing, migration checklists, COST_FIRST/SPEED_FIRST/SUCCESS_FIRST routing, and querying current availability/pricing snapshots before discussing specific models. In contrast, this file merely collects CLI arguments and writes a deterministic draft blueprint JSON without implementing any auditing, pricing lookup, compatibility analysis, or runnable migration assistance.

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:

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
97% confidence
Finding

The module docstring and user-facing instructions are written entirely in Chinese, and subsequent prompts/messages throughout the script continue that assumption. There is no indication that the skill is region-specific or that users can opt into another language, which creates a natural-language locale policy concern.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

Medium
Category
Not specified by scanner
Confidence
90% confidence
Finding

The manifest says this skill should help users judge whether AI-HIVE is a suitable fal.ai alternative by delivering requirements inventory, compatibility audit, migration checklist, routing guidance, runnable code, task ledger, and acceptance criteria. In contrast, the code is an operational AI-Hive client that invokes chat, image generation, video generation, upload, and polling endpoints directly, with no implementation of comparison/audit/migration logic. This is a semantic mismatch between the described advisory workflow and the actual executable behavior.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The wrapper is presented as an image-generation skill, but it also exposes generic task lookup and arbitrary file upload operations. In an agent/skill environment, this expands capability beyond the declared purpose and can enable unintended data exfiltration, probing of task metadata, or misuse of the skill as a general media transport primitive.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
97% confidence
Finding

The generated audit checklist strings are entirely in Chinese, which makes the skill's natural-language output language-specific by default. The file provides no option for the user to choose language and no justification that this is a region-specific or Chinese-only compliance tool.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The file is packaged as a skill for fal.ai alternative evaluation and video-generation-oriented assistance, but it exposes a much broader generic AI API client surface including chat, image generation, model enumeration, uploads, downloads, and account operations. This scope expansion increases the available capability set beyond what users would reasonably expect from the skill metadata, creating an over-privileged tool that could be repurposed for unintended actions or data handling.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
98% confidence
Finding

The module docstring and all user-facing CLI guidance are written in Chinese, presenting the skill as a general-purpose tool rather than a region-specific or Chinese-only skill. This creates a language policy concern because users are not given any language/locale opt-in or alternative.

Content

No source excerpt is available for this finding.

Context-Inappropriate Capability

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The skill includes a user-info endpoint that retrieves account-level information and wallet balance, which is not necessary for the stated purpose of migration/comparison and video generation. Unnecessary access to billing/account data expands the sensitivity of accessible information and could enable privacy leakage or account reconnaissance if the skill is invoked in broader agent workflows.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The upload flow sends local media files to remote API-controlled storage and later downloads result URLs, yet there is no explicit warning, confirmation, or destination transparency for users at the point of transmission. In a skill context handling potentially sensitive creative assets, this creates a meaningful data exposure risk because users may not realize files leave the local environment and are stored by third-party infrastructure.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
88% confidence
Finding

This Python file contains user-facing natural-language strings such as the module docstring, CLI description, skill name, default platform value, and search aliases in Chinese, but provides no opt-in or alternative locale selection. Under the policy rule, forcing a specific language without user choice is a natural-language policy concern even though the code itself is otherwise simple.

Content

No source excerpt is available for this finding.

Intent-Code Divergence

Low
Category
Not specified by scanner
Confidence
83% confidence
Finding

The top-level docstring states this is a universal AI Hive tool covering chat, image, video, upload, and polling. However, the executable entrypoint and embedded skill configuration package it as a specific video-generation skill (videogen.py) with constrained mode selection and a concrete skill identity, creating active documentation-level divergence about what this file is intended to be.

Content

No source excerpt is available for this finding.

Context-Inappropriate Capability

Low
Category
Not specified by scanner
Confidence
79% confidence
Finding

The init flow opens a browser URL containing a skill-identifying query parameter for attribution. While not directly code-execution dangerous, it transmits usage metadata that is not essential to obtaining an API key and can create unnecessary tracking of which skill initiated the workflow.

Content

No source excerpt is available for this finding.