ClawHub
Back to skill

Security audit

seedance-combat-director

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed Chinese-language prompt-writing workflow for Seedance action videos, with no evidence of hidden access, persistence, exfiltration, or destructive behavior.

Install only if you want a Chinese-first Seedance combat prompt director. Provide only the videos, images, and brief details you want analyzed, and ask explicitly if you want output in another language or format.

A.I.G

Vulnerability Patterns
  • 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
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs

SkillSpector

Vulnerability Patterns
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • 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
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (4)

Natural-Language Policy Violations

Medium
Confidence
95% confidence
Finding
Line L121 says the final output is a copyable Chinese `text` code block, which imposes a specific language as part of the workflow. The README does not indicate that users may opt into another language or choose locale preferences, so this is a natural-language policy concern.

Lp3

Medium
Category
MCP Least Privilege
Confidence
93% confidence
Finding
The skill explicitly instructs the agent to read local reference files and, in some workflows, to inspect provided videos before proceeding, but it declares no tool scope or permissions boundary. That mismatch can lead to overbroad file-read capability exposure, making unintended access to unrelated local files or artifacts more likely if the runtime grants generic file access.

Natural-Language Policy Violations

Medium
Confidence
94% confidence
Finding
The markdown states that for complete new tasks the skill should output a directly copyable Chinese `text` code block by default. This imposes a specific language choice unless the user explicitly requests otherwise, which is a natural-language locale policy issue because it does not default to user choice or neutral language handling.

Natural-Language Policy Violations

Medium
Confidence
92% confidence
Finding
The entire template content is written in Chinese and includes a user-facing note in Chinese, which implies the skill expects Chinese-language use. The file does not offer any language or locale choice, nor does it document that this is a region-specific or language-specific tool, which can violate language/locale policy requirements.

Static analysis

No suspicious patterns detected.