ClawHub

Instagram API

Security checks across malware telemetry and agentic risk

Overview

This is a documented Instagram data API helper whose main risk is ordinary third-party API credential and query sharing with agntdata.

Install only if you trust agntdata with your Instagram research targets and API credential. Keep AGNTDATA_API_KEY in an environment variable, avoid putting secrets or confidential client details in prompts, logs, or the useCase field, monitor credit usage, and review the separate plugin package before choosing the plugin route.

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 (3)

Intent-Code Divergence

Low
Confidence
96% confidence
Finding
The skill documents Bearer-token authentication in the main authentication section, but later provides an example using a different header (`X-API-Key`). Conflicting auth guidance can cause agents or users to send credentials incorrectly, leading to failed requests, accidental credential exposure in debugging, or implementation of insecure fallback behavior.

Missing User Warnings

Medium
Confidence
90% confidence
Finding
The skill promotes sending Instagram queries, usernames, IDs, URLs, and an API credential to a third-party service without a clear privacy or data-handling warning. In an agent setting, this increases the risk that users unknowingly transmit sensitive business research, personal identifiers, or account-related data off-platform.

Missing User Warnings

Medium
Confidence
94% confidence
Finding
The API key activation step sends both the bearer credential and a free-form `useCase` string to an external registration endpoint, but the instructions do not warn users that this metadata leaves the local environment. That omission can lead users or autonomous agents to disclose internal project details, client names, or other sensitive operational context during setup.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal