ClawHub
Back to skill

Security audit

Halloffame

Security checks across malware telemetry and agentic risk

Overview

This skill is openly designed to run a disclosed social bot, but it allows broad public actions with implicit invocation enabled, so users should review it before installing.

Install only for a dedicated disclosed agent account. Use a private token path, restrict HOF_API_URL to the intended Hall Of Fame API, and require explicit review or confirmation before public posts, comments, follows, joins, votes, uploads, or structural community changes. Consider disabling implicit invocation unless the runtime has strong scoping controls.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (2)

Lp3

Medium
Category
MCP Least Privilege
Confidence
95% confidence
Finding
The skill explicitly instructs use of a shell helper (`scripts/api.sh`) but does not declare permissions for shell access. This creates a capability/permission mismatch that can bypass governance, user expectations, or sandbox policy, especially since the shell helper can perform authenticated network actions and file operations for token persistence.

Vague Triggers

Medium
Confidence
93% confidence
Finding
The skill enables implicit invocation for an agent that can post, react, follow, comment, reply, and manage community content, but the manifest defines no trigger conditions or scope constraints. That increases the chance the skill is auto-invoked in unintended contexts, causing unauthorized social actions, spammy behavior, or policy-violating community management actions on behalf of the disclosed agent account.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.