ClawHub
Back to skill

Security audit

LOFTER

Security checks for vulnerabilities and agentic risk

Overview

The skill is only a Markdown guide, but its LOFTER community description conflicts with streaming-style instructions and broad example triggers.

Review this skill before installing because its description and instructions do not describe the same capability set. It does not appear to run code or access data, but it should be corrected to focus on LOFTER community and creator guidance, with narrower examples and clear language scope.

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
  • 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
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (4)

Description-Behavior Mismatch

Medium
Category
Not specified by scanner
Confidence
97% confidence
Finding

The skill manifest says LOFTER provides community and creative-guidance content, but the documentation describes video-streaming features such as playback controls, membership status, downloads, and bitrate-related options. This mismatch can cause the agent to route unrelated user requests to the skill or misrepresent capabilities, increasing the risk of user deception, unsafe action selection, and policy bypass through disguised scope.

Content

No source excerpt is available for this finding.

Intent-Code Divergence

Medium
Category
Not specified by scanner
Confidence
98% confidence
Finding

The file documents a different product/function set than the declared skill, creating a scope-identity contradiction. In agent ecosystems, this is dangerous because skill selection and trust often rely on metadata and documentation consistency; contradictory docs can be used to capture prompts outside intended scope and confuse downstream reviewers or users.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The example trigger phrases are highly generic, such as asking how to use LOFTER for a 'specific scenario task' or asking about a 'core function,' which could match many ordinary requests. Overbroad triggers increase the chance of inappropriate invocation, prompt hijacking of general conversations, and user confusion about what the skill is actually authorized to do.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
88% confidence
Finding

All user-facing description and examples are presented only in Chinese, and the file does not indicate that the user can choose another language or that the skill is intentionally limited to a Chinese-language audience for a stated reason. This can violate language or locale policy when a skill implicitly forces a specific language without user opt-in.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.