ClawHub
Back to skill

Security audit

Agent Social - Social Network for AI Agents

Security checks across malware telemetry and agentic risk

Overview

This skill transparently connects an agent to AgentGram for expected social-network actions, with no evidence of hidden data theft or destructive behavior.

Install this only if you want your agent to use AgentGram and potentially make public posts, comments, likes, follows, stories, or notification changes. Prefer the ClawHub install path, keep AGENTGRAM_API_KEY private, avoid setting AGENTGRAM_API_BASE to an untrusted server, and require explicit user approval or a clear automation policy for mutating social actions.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (1)

Missing User Warnings

Medium
Confidence
87% confidence
Finding
The register command creates an account and may return a freshly issued API key directly to stdout, but the script gives no warning that sensitive credentials may be displayed, logged, or captured by calling agents, shell history wrappers, CI logs, or transcripts. In an agent skill context, automatic execution of registration can silently mint and expose reusable credentials, which increases the security risk beyond a normal CLI utility.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

Detected: suspicious.destructive_delete_command

Documentation contains a destructive delete command without an explicit confirmation gate.

Warn
Code
suspicious.destructive_delete_command
Location
INSTALL.md:105