ClawHub
Back to skill

Security audit

Runway 视频生成替代｜AI 视频生成与编辑

Security checks for vulnerabilities and agentic risk

Overview

This appears to be a legitimate AI Hive video-generation skill, but its helper script moves user media through server-provided upload and download URLs without enough destination or size controls.

Review this skill before installing if you will use private, customer, or unreleased media. Prefer running it in a constrained environment, use a limited AI Hive API key, verify the service domain behavior, and be aware that generated results are downloaded locally by default.

A.I.G

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)

T08 · Insecure Dependencies

Note
Location
SKILL.md:53
Finding

Unpinned Third-Party Runtime Dependency

Content
View full analysis
Remediation
View remediation
--hash=sha256: ``` 4. Install dependencies using hash enforcement: ```bash python3 -m pip install --require-hashes -r requirements.txt ``` 5. Use an isolated virtual environment rather than modifying the global Python environment. 6. Add automated dependency vulnerability scanning and a controlled process for reviewing and updating pinned versions. ]]>

T09 · Insecure Skill Coding Practices

Warning
Location
scripts/videogen.py:229
Finding

Unvalidated API-Supplied Upload and Download Destinations

Content
View full analysis
0: pct = int(downloaded * 100 / total) bar = "█" * (pct // 5) + "░" * (20 - pct // 5) print(f"\r {bar} {pct}% ({downloaded // 1024}KB)", end="", flush=True) ``` ### Technical Analysis The program trusts `upload.url`, `upload.method`, upload headers, `resultUrl`, and `lastFrameUrl` returned by the AI Hive API. It does not enforce HTTPS, allowlist expected object-storage domains, reject loopback or private network addre ...[truncated 2382 chars]
Remediation
View remediation

SkillSpector

Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (4)

Undeclared Tool Scope

Medium
Category
MCP Least Privilege
Confidence
91% confidence
Finding

The skill invokes a Python script with network, shell, file read/write, and likely environment access, but the manifest does not declare any tool scope or permission boundaries. This is dangerous because an agent may execute the skill with broader-than-expected capabilities, increasing the blast radius for prompt injection, unsafe file access, secret exposure, or unintended outbound requests.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The description contains broad trigger phrases such as Runway alternative, API, video generation, editing, extension, and ad/video use cases, which can cause the skill to activate for a wide range of common requests. Overbroad activation is risky because users may be routed into a skill that performs shell and network-backed operations without clearly intending to use this specific workflow.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The module docstring and all user-facing CLI guidance are written only in Chinese, and the skill title/configuration also targets Chinese-language usage. This imposes a language choice on users without any explicit opt-in or alternative locale support, which matches the language/locale policy violation criteria.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
83% confidence
Finding

The file content and user-facing description are primarily in Chinese, with embedded English trigger terms, but there is no statement that the skill can adapt to the user's preferred language. Under the policy, forcing or assuming a language without opt-in can be a natural-language policy issue.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.