ClawHub

Instagram Skill via cyberdrk/gram CLI

Security checks across malware telemetry and agentic risk

Overview

This is a disclosed Instagram CLI skill that uses session cookies and can change account state, so it is acceptable but should be used carefully.

Install only if you trust the gram npm package with your Instagram session. Treat sessionid, csrftoken, ds_user_id, and browser cookie profiles like passwords, avoid saving them in shared config or shell history, and require an explicit user request before running like, comment, save, follow, unfollow, or similar account-changing commands.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • 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)

Missing User Warnings

Medium
Confidence
92% confidence
Finding
The skill instructs users to supply or extract Instagram authentication cookies (`sessionid`, `csrftoken`, `ds_user_id`) but does not clearly warn that these are live account credentials equivalent to session secrets. In a skill context, encouraging browser-cookie sourcing without an explicit security warning increases the risk of accidental credential exposure, unsafe sharing, or use on untrusted systems.

Missing User Warnings

Medium
Confidence
90% confidence
Finding
The documented engagement commands (`like`, `comment`, `follow`, etc.) perform real account-modifying actions on Instagram, but the skill does not prominently warn users before presenting them as ordinary read operations. This can lead to unintended actions, account abuse, rate-limit issues, or reputational harm if users run commands without understanding they change remote state.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal