ClawHub
Back to skill

Security audit

Seedance 2.5 提示词专家｜AI-HIVE

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed AI-HIVE video workflow helper that can upload selected media and make billable API calls, with instructions to confirm before generation.

Install only if you intend to use AI-HIVE for Seedance video workflows. Review prompts and routing before generation, avoid uploading media you are not authorized to use, and keep the AI-HIVE API key out of logs, screenshots, and repositories.

SkillSpector

Vulnerability Patterns
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (3)

Lp3

Medium
Category
MCP Least Privilege
Confidence
93% confidence
Finding
The skill includes executable code examples and operational guidance that use environment variables, local files, network access, and shell commands, but it does not declare any corresponding permissions. This creates a transparency and policy-enforcement gap: an agent or user may invoke capabilities with external side effects, including billable API calls, file access, and data transmission, without an explicit permission boundary.

Vague Triggers

Medium
Confidence
92% confidence
Finding
The manifest enables implicit invocation (`allow_implicit_invocation: true`) without any visible guardrails in this file that restrict activation to narrowly scoped user intents. That can cause the skill to trigger in broader contexts than intended, exposing users to unexpected workflow execution, prompt shaping, or external API-oriented behavior without explicit consent.

Natural-Language Policy Violations

Medium
Confidence
84% confidence
Finding
The default prompt hard-codes 'production-ready Chinese workflow' rather than respecting user language preference or requiring opt-in. While not a direct code-execution issue, this can override user intent, create misleading outputs, and increase the chance that the skill is invoked or behaves unexpectedly in contexts where language choice matters for safety, review, or downstream automation.

Static analysis

No suspicious patterns detected.