ClawHub
Back to skill

Security audit

Instagram Publisher

Security checks for vulnerabilities and agentic risk

Overview

This skill openly helps publish and schedule Instagram content through a third-party MCP credential, with no local executable code, but users should treat the connector link as account write access.

Install only if you trust Boring with publish access to your connected Instagram Business or Creator account. Keep the MCP URL private like a password, review posts before publishing or scheduling, and revoke or regenerate the connector if it is exposed or no longer needed.

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
  • 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)

Intent-Code Divergence

Medium
Category
Not specified by scanner
Confidence
98% confidence
Finding

The skill’s Security & Data Handling section contains a misleading assurance that platform interactions only retrieve analytics and do not upload or modify content, while the rest of the skill explicitly supports file upload, media re-hosting, publishing, scheduling, and cancellation of posts. This inconsistency can cause users or downstream agents to underestimate the write-capable scope of the embedded MCP credential and approve actions or installations under false assumptions.

Content

No source excerpt is available for this finding.