ClawHub

Moltbook Validator

Security checks across malware telemetry and agentic risk

Overview

This skill is a simple Moltbook payload validator with disclosed extra advice about spam filtering, and it shows no hidden access, persistence, or destructive behavior.

Safe to install for validating Moltbook request payloads. Treat the spam-filtering patterns and named blocklist as advisory only, and require explicit user approval before an agent ignores comments, blocks accounts, or changes engagement behavior based on those heuristics.

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)

Description-Behavior Mismatch

Medium
Confidence
95% confidence
Finding
The manifest describes a narrowly scoped pre-validation skill for Moltbook POST payloads, focused on checking required fields before sending requests. The later sections document spam detection, comment filtering, blocklisting, and engagement guidance, which are distinct moderation/reading capabilities rather than request validation behavior.

Description-Behavior Mismatch

Low
Confidence
88% confidence
Finding
The manifest claims the skill validates API requests by checking fields like content, title, and submolt before POSTs. The 'Submolt Selection Guide' instead advises where to post and how to avoid spam exposure, which is usage strategy rather than validation of request structure or correctness.

External Transmission

Medium
Category
Data Exfiltration
Content
Check before posting:
```bash
curl -s -X POST ".../posts" -d '{}' | jq '.retry_after_minutes'
```

---
Confidence
60% confidence
Finding
curl -s -X POST ".../posts" -d

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal