Back to skill

Security audit

自然人像精修 · Portrait Retouch

Security checks for vulnerabilities and agentic risk

Overview

The verdict is benign because the reported behavior appears purpose-aligned, though the scanners flag an install command that retrieves a skill and note incomplete inspection of referenced images.

Before installing, consider that only half of the components were fully scanned and the three referenced images were not fully interpreted. The scan also flags the unpinned installer and says the installed skill's implementation could not be verified, so review the source and pin reviewed versions if you proceed.

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
Vulnerability Patterns
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • 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 (5)

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
97% confidence
Finding

The supplied command performs skill installation, not photo retouching. Although the installed skill's name matches the declared purpose, its implementation is not supplied and cannot be verified. The visible behavior is therefore materially different from the described workflow; no additional behavior of the external skill is inferred.

Content

No source excerpt is available for this finding.

Incomplete referenced artifact analysis

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 30)May include surrounding context.

md
[示例原图](assets/before.png)是 AI 生成的虚构人像。[精修成片](assets/after.png)由该原图编辑得到。[查看并排对比](assets/comparison.jpg)。

Incomplete referenced artifact analysis

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 30)May include surrounding context.

md
[示例原图](assets/before.png)是 AI 生成的虚构人像。[精修成片](assets/after.png)由该原图编辑得到。[查看并排对比](assets/comparison.jpg)。

Incomplete referenced artifact analysis

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 30)May include surrounding context.

md
[示例原图](assets/before.png)是 AI 生成的虚构人像。[精修成片](assets/after.png)由该原图编辑得到。[查看并排对比](assets/comparison.jpg)。

MCP server referenced without pinned version: 'npx skills'.

Medium
Category
MCP Rug Pull
Confidence
92% confidence
Finding

The installation example uses an unpinned npm CLI, allowing npx to download and execute a version that can change over time. Although this is a Skills CLI rather than an MCP server, a compromised package release could execute code with the installing user's privileges; this is a supply-chain risk, not evidence of malicious retouching behavior.

Content

No source excerpt is available for this finding.