Back to skill

Security audit

de-bloat 至简

Security checks for vulnerabilities and agentic risk

Overview

ClawScan rates this skill benign with medium confidence: scanners found no direct malicious behavior, while noting broad activation wording and purpose mismatches in evaluation samples; coverage was partial.

The verdict is benign, but treat it as a cautious assessment rather than a guarantee: Skillspector inspected 97.2% of components, with one component partially inspected and a static tool-misuse analyzer degraded. Before installing, consider whether broad phrase-based activation fits your use and whether you are comfortable with the unresolved purpose mismatch flags in the evaluation samples.

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
  • 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 (3)

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
94% confidence
Finding

The description presents a general code-development and simplification skill, whereas the supplied code implements a specific coupon-format validator. It contains no behavior implementing the described development workflow. No external resource access, side effects, or suspicious instructions are present.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
92% confidence
Finding

The supplied code implements application-specific SKU validation rather than coding assistance or code simplification. Its use of JavaScript alone does not establish alignment with the declared skill. No file, network, or other external resource access is present.

Content

No source excerpt is available for this finding.

Common phrases can activate the skill without an explicit code-change requirement.

Medium
Category
Trigger Abuse
Confidence
95% confidence
Finding

L003 lists everyday phrases such as 「简单点」 ('make it simpler') as sufficient activation conditions and says to use the skill even when it is not named. Although the description opens with coding tasks, it does not explicitly restrict these phrase-based triggers to code creation or modification, or exclude explanation-only requests, so unrelated simplification requests could invoke the coding workflow.

Content

No source excerpt is available for this finding.