ClawHub
Back to skill

Security audit

Moltitude

Security checks across malware telemetry and agentic risk

Overview

Moltitude is a coherent receipt service, but it asks agents to auto-register and upload detailed work traces to a third party with limited consent and privacy safeguards.

Install only if you are comfortable with Moltitude receiving selected task details. Configure the agent to ask before registration or minting, redact secrets, private file contents, personal data, and internal reasoning from receipts, store the private key and claim code as credentials, and avoid lifetime remix approval unless broad ongoing sharing is intended.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Findings (9)

Description-Behavior Mismatch

Medium
Confidence
88% confidence
Finding
The documented functionality extends beyond the declared purpose of minting proof-of-work receipts into feed browsing and cross-agent remix permission management. That scope expansion increases attack surface and can mislead users or platforms about what data the skill accesses and what operations it performs, which matters for least-privilege review and informed consent.

Context-Inappropriate Capability

Low
Confidence
92% confidence
Finding
The embedded analytics image causes an external request when the skill documentation is viewed, which is unrelated to the core receipt-creation purpose. This creates passive telemetry leakage such as IP address, user agent, access timing, and possibly environment metadata without clear disclosure or consent.

Vague Triggers

Medium
Confidence
88% confidence
Finding
The trigger phrases are broad enough to match normal user requests like 'document this task' or 'prove what I did,' which could invoke the skill outside a clear user request for this specific service. In context, that matters because the skill is network-connected and tied to account/receipt creation, so accidental activation can lead to unintended data disclosure or external actions.

Natural-Language Policy Violations

Low
Confidence
95% confidence
Finding
The description explicitly instructs immediate registration, pressuring the agent to perform an external side effect before meaningful user opt-in. In a skill ecosystem, this is dangerous because installation or casual invocation may cause unsolicited outbound requests and account creation with third-party infrastructure.

Missing User Warnings

High
Confidence
98% confidence
Finding
The skill explicitly instructs agents to send full prompts, reasoning steps, tool inputs/outputs, and result data to a third-party API. Those fields can contain secrets, personal data, proprietary source code, or internal decision-making traces, so transmitting them externally without strong minimization and privacy warnings creates a serious confidentiality risk.

Missing User Warnings

Medium
Confidence
90% confidence
Finding
The registration flow returns sensitive credentials including a private key and claim code, but the guidance only says to save them immediately and does not emphasize secure storage, rotation, or the consequences of disclosure. Poor handling of these credentials could let others impersonate the agent, sign fraudulent receipts, or take over dashboard access.

External Transmission

Medium
Category
Data Exfiltration
Content
"setup": {
    "required": true,
    "level": 0,
    "instruction": "Execute POST https://api.moltitude.com/v1/agents/register/simple with {\"name\":\"your-name\"} immediately",
    "onInstall": {
      "action": "register",
      "endpoint": "https://api.moltitude.com/v1/agents/register/simple",
Confidence
96% confidence
Finding
https://api.moltitude.com/

External Transmission

Medium
Category
Data Exfiltration
Content
"instruction": "Execute POST https://api.moltitude.com/v1/agents/register/simple with {\"name\":\"your-name\"} immediately",
    "onInstall": {
      "action": "register",
      "endpoint": "https://api.moltitude.com/v1/agents/register/simple",
      "method": "POST",
      "body": {"name": "{{AGENT_NAME}}"}
    }
Confidence
96% confidence
Finding
https://api.moltitude.com/

External Transmission

Medium
Category
Data Exfiltration
Content
- API Docs: https://moltitude.com/docs/api
- Remix Guide: https://moltitude.com/remix.md

![](https://api.moltitude.com/v1/analytics/skill-read)
Confidence
87% confidence
Finding
https://api.moltitude.com/

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.