T09 · Insecure Skill Coding Practices
Error
- Location
main.py:14- Finding
Unbounded Image Decoding Enables Resource Exhaustion
- Content
View full analysis
- Remediation
View remediation
Security audit
图片智能去重 - 递归全目录·安全回收站版
Security checks for vulnerabilities and agentic risk
Overview
This image deduplication skill does what it claims, but it can automatically move many user files with overstated safety claims and no review or rollback step.
Install only if you are comfortable with an automated tool moving images out of their original folders. Run it first on a small copied test folder, use a very strict similarity value, and avoid pointing it at a whole home directory, shared business library, or irreplaceable photo archive until it has a dry-run or review mode.
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 (3)
T09 · Insecure Skill Coding Practices
Warning
- Location
main.py:75- Finding
Automatic Bulk File Relocation Without Confirmation or Rollback
- Content
View full analysis
- Remediation
View remediation
other
Note
- Location
README.md:26- Finding
Documentation Misrepresents Local Relocation as Risk-Free Trash Handling
- Content
View full analysis
- Remediation
View remediation
Vulnerability Patterns
- Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
- MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
- 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 (4)
Undeclared Tool Scope
Medium
- Category
- MCP Least Privilege
- Confidence
- 89% confidence
- Finding
The skill advertises recursive scanning of arbitrary folders and necessarily reads many files, but it does not declare any explicit tool scope or permissions boundaries. That increases the risk of over-broad file access because an agent may invoke it against sensitive directories without clear least-privilege constraints or user-visible permission metadata.
- Content
Vague Triggers
Medium
- Category
- Not specified by scanner
- Confidence
- 93% confidence
- Finding
The invocation phrases are broad natural-language patterns such as 'scan all images in [folder path] with [similarity] deduplication,' which could match normal user requests and trigger the skill unexpectedly. Because this skill performs recursive file operations and moves files to trash, accidental invocation could cause unintended large-scale file enumeration or modifications.
- Content
Natural-Language Policy Violations
Low
- Category
- Not specified by scanner
- Confidence
- 69% confidence
- Finding
The document is presented only in Chinese and English, which may imply a fixed language set rather than offering user choice. If the skill itself is intended to operate only in these languages, that should be explicitly documented as a justified constraint or presented as an option rather than an assumed default.
- Content
Missing User Warnings
Low
- Category
- Not specified by scanner
- Confidence
- 88% confidence
- Finding
The markdown states that the tool will automatically move low-quality duplicate images to the recycle bin/trash, which is a file-modifying operation that can affect user data organization. Although it says files are not deleted, it does not clearly warn users about the operational impact or recommend reviewing results before running it.
- Content