ClawHub
Back to skill

Security audit

MemData

Security checks across malware telemetry and agentic risk

Overview

MemData is a disclosed remote memory service, but users should be careful about storing sensitive data and approving wallet-paid actions.

Install only if you are comfortable with a remote service storing agent memories under a wallet identity and charging per API call. Use a limited wallet, review payment/signature prompts, avoid storing sensitive data in standard mode, enable encryption for private memories, and require explicit confirmation before deleting stored memories.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Tool MisuseTool Parameter Abuse, Chaining Abuse, Unsafe Defaults
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (3)

Missing User Warnings

Medium
Confidence
90% confidence
Finding
The skill promotes persistent memory tied to a wallet identity and cross-session retention, but it does not clearly warn users that data may persist indefinitely, be correlated across sessions by wallet address, and in standard mode be readable by the service. In an agent context, this can lead to accidental storage of sensitive prompts, business data, or personal information without informed consent.

Missing User Warnings

Medium
Confidence
94% confidence
Finding
The documented delete endpoint allows removal of a memory and all associated chunks, but there is no warning that deletion may be irreversible or that agents should confirm intent before calling it. In autonomous-agent workflows, an LLM or indirect prompt could trigger destructive deletion and cause permanent loss of memory state.

Tool Parameter Abuse

High
Category
Tool Misuse
Content
---

### DELETE /artifacts/:id

Delete a memory and all its chunks.
Confidence
91% confidence
Finding
DELETE /artifacts/:id

VirusTotal

59/59 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.