ClawHub
Back to skill

Security audit

Alibaba Cloud AI Video LivePortrait

Security checks for vulnerabilities and agentic risk

Overview

This skill is a narrow Alibaba Cloud LivePortrait helper, with disclosed but privacy-sensitive use of public portrait and audio URLs.

Install only if you are comfortable sending portrait images and voice audio to Alibaba Cloud Model Studio and making those inputs reachable by public HTTP/HTTPS URL. Use short-lived links where possible, avoid embedding long-lived secrets in URLs, and clean up output/aliyun-liveportrait if it contains sensitive request records.

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
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Rogue AgentSelf-Modification, Session Persistence
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (4)

Lp3

Medium
Category
MCP Least Privilege
Confidence
94% confidence
Finding
The skill declares network and file-write behaviors but does not define an explicit tool scope such as allowed-tools or permissions. That omission weakens least-privilege boundaries and can let an agent invoke broader capabilities than users expect, especially since the workflow writes artifacts and uses public URLs for remote resources.

Session Persistence

Medium
Category
Rogue Agent
Content
## Validation

```bash
mkdir -p output/aliyun-liveportrait
python -m py_compile skills/ai/video/aliyun-liveportrait/scripts/prepare_liveportrait_request.py && echo "py_compile_ok" > output/aliyun-liveportrait/validate.txt
```
Confidence
60% confidence
Finding
Skill establishes unauthorized persistence across sessions via cron jobs, startup scripts, or state files. Session persistence allows an attacker to maintain access beyond the current interaction.

Missing User Warnings

Medium
Confidence
97% confidence
Finding
The skill explicitly instructs operators to save exact portrait/audio URLs and task polling snapshots, which can retain sensitive media locations, tokens embedded in URLs, or other request metadata. Persistent logging of these artifacts increases the exposure window for personal data and may leak access credentials if URLs are signed or query-parameterized.

Missing User Warnings

Medium
Confidence
96% confidence
Finding
Requiring public image and audio URLs means potentially sensitive portrait and voice data must be exposed on the public internet to use the skill. Combined with no privacy warning or handling guidance, this creates a material risk of unintended disclosure, scraping, and long-lived access to biometric or personal media.

Static analysis

No suspicious patterns detected.