Back to skill

Security audit

Taplio LinkedIn Comment Reply Manager

Security checks for vulnerabilities and agentic risk

Overview

ClawScan rates this skill suspicious because it may queue public comment likes under approval intended for replies, without clearly disclosed consent for likes.

Before installing, consider whether you are comfortable with the skill potentially liking comments from your account when approving replies. The scan found no executable scripts and AIG reported no findings, but Skillspector marked coverage partial because a local reference was missing, so the evidence is incomplete.

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
  • 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
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (1)

The skill can queue comment likes in addition to the approved replies described in its manifest.

Medium
Category
Excessive Agency
Confidence
94% confidence
Finding

The skill permits queuing comment likes on the same approval used to commit replies, although the review does not explicitly request approval for likes. This can create public endorsements from the user's account without informed authorization; limiting likes to comments being answered reduces the scope but does not establish consent.

Content

No source excerpt is available for this finding.