ClawHub
Back to skill

Security audit

Social Publisher

Security checks across malware telemetry and agentic risk

Overview

This is a legitimate-looking social media automation skill, but it asks for account tokens and can make public posts and engagement actions without enough guardrails.

Review before installing. Use limited-scope tokens, keep the config file out of repositories and shared folders, restrict file permissions, and start only in preview or draft mode. Require human approval for generated replies, follows, reposts, and cross-platform scheduled posts, and verify how to pause or stop workflows before connecting real accounts.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • 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
90% confidence
Finding
The skill explicitly promotes autonomous public interactions such as liking, retweeting, replying, and follow-backs on connected social accounts without any warning, approval gate, or abuse safeguards. In context, this increases the risk of account misuse, spammy or policy-violating behavior, reputational damage, and unintended public statements made on behalf of the user or organization.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The setup instructions place API keys, client secrets, and access tokens directly into a local YAML file without any credential-handling guidance. This encourages insecure secret storage, increasing the chance of accidental commit to source control, leakage through logs or backups, and compromise of connected social-media accounts and associated business assets.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.