ClawHub
Back to skill

Security audit

partial-redrawing

Security checks for vulnerabilities and agentic risk

Overview

This image-editing skill is review-worthy because local images may be uploaded to Flyelep and exposed through permanent public links without a clear user-facing consent step or deletion path.

Only install this if users understand that local images supplied for editing can be sent to Flyelep and may become accessible through permanent public links. Avoid using it for identity documents, private photos, confidential designs, regulated data, or other sensitive images unless the user explicitly accepts that exposure. The API key should be provided only at runtime and not saved in files or shared logs.

A.I.G

Vulnerability Patterns
  • 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
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
Findings (1)

T09 · Insecure Skill Coding Practices

Warning
Location
SKILL.md:112
Finding

Uploaded Images Are Exposed Through Permanent Unsigned Public URLs

Content
View full analysis
Remediation
View remediation

SkillSpector

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (4)

Missing User Warnings

High
Category
Not specified by scanner
Confidence
97% confidence
Finding

The skill instructs the agent to upload user-provided local images to a permanent public URL and explicitly states the resulting link is permanently valid, but it does not require a clear consent and privacy warning before doing so. This can expose sensitive personal, proprietary, or regulated image content to public internet access without the user fully understanding the permanence and visibility of the upload.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
84% confidence
Finding

The skill description and operational instructions are entirely in Chinese, and the examples direct prompt content in Chinese without indicating that users may choose another language. Because the file includes a languageType parameter but does not offer opt-in language selection or explain a justified locale restriction, it presents a natural-language locale constraint without user choice.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The invocation text says to use this skill whenever the user asks to partially modify an image, replace a background, replace some region content, or keep the subject while changing part of the image. These are broad, common phrasings in everyday image-editing requests, and the file does not provide explicit trigger boundaries or negative examples to clarify when this skill should not activate.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
92% confidence
Finding

The skill directs external transmission of local user files to a third-party service, which is expected for the feature, but it remains security-relevant because sensitive local content may leave the user's environment. In this skill's context, the danger is increased by the same section noting the upload yields a permanent public URL, compounding confidentiality and privacy exposure.

Content

Scanner excerpt · SKILL.md (reported line 140)May include surrounding context.

判断成功只看 code，业务失败时 HTTP 状态码仍是 200，code 为 500 或 9999，原因在 msg 里。

bash
# Windows/PowerShell（用 curl.exe，PowerShell 里的 curl 是 Invoke-WebRequest 的别名）
curl.exe -X POST "https://www.flyelep.cn/prod-api/poster-design/api/v1/file/upload" -H "secretKey: 你的密钥" --max-time 120 -F "file=@C:/path/to/product.png"

# macOS/Linux

Static analysis

No suspicious patterns detected.