Unipile Instagram Sdk

Security checks across malware telemetry and agentic risk

Overview

The skill matches its advertised Instagram automation purpose, but it enables write-capable account actions by default while also claiming read-only safety.

Review this before installing. Use it only if you trust Unipile with the connected Instagram account, set UNIPILE_PERMISSIONS=read unless you intentionally need posting or messaging, and consider a dedicated account rather than a personal main account.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

Intent-Code Divergence

Medium
Confidence
97% confidence
Finding
The documentation is internally inconsistent: it states the skill is read-only by default, but elsewhere documents the default as `read,write`. For a skill that can send DMs, create posts, and react/comment on behalf of a user, this mismatch can cause users to believe they are operating safely while the skill actually has write capability.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The CLI defaults to `read,write` permissions when `UNIPILE_PERMISSIONS` is unset, which means operators may unknowingly run a tool capable of sending messages, starting chats, commenting, reacting, or creating posts. In an agent/automation context, insecure defaults increase the chance of unintended outbound actions on a real Instagram account and violate least-privilege expectations.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal