ClawHub

memic

Security checks across malware telemetry and agentic risk

Overview

This is a documentation-only Memic SDK skill whose sensitive data flows are mostly expected for a managed RAG and Text2SQL service.

Before installing, verify the memic package source, pin versions for production, use a scoped Memic API key, avoid uploading secrets or regulated data without approval, and use read-only least-privilege database credentials or non-production replicas when configuring connectors.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (3)

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The skill explicitly promotes uploading documents and using search/chat against a managed external SaaS, but it does not clearly warn users that document contents, prompts, and retrieved context may be transmitted to and stored/processed by a third party. In an agent setting, users may pass proprietary, regulated, or personal data assuming the tool is local, which creates confidentiality and compliance risk.

Missing User Warnings

Medium
Confidence
97% confidence
Finding
The database connector guidance tells users to enter PostgreSQL/MySQL connection details into the service but omits an explicit warning that credentials, schema metadata, natural-language queries, generated SQL, and query results may be exposed to and processed by a third-party platform. This is more dangerous than document upload because it can enable broad access to live structured data and potentially sensitive production records.

Missing User Warnings

Medium
Confidence
90% confidence
Finding
The examples demonstrate uploading local files and sending user queries and retrieved document content to a remote Memic service, but they provide no notice about data transmission, privacy implications, or suitability for sensitive data. In a context-engineering/RAG SDK, this omission can cause developers to unknowingly exfiltrate confidential documents, database-derived context, or regulated data to a third-party service and then onward into LLM prompts.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal