ClawHub

Datagma

Security checks across malware telemetry and agentic risk

Overview

This is a disclosed Datagma integration that can perform sensitive people and contact enrichment, so it is usable but should be handled with privacy care.

Install only if you trust Membrane and Datagma and need enrichment workflows. Require explicit confirmation before person-level lookups or raw proxy requests, and avoid sending personal identifiers unless you have a lawful, policy-compliant reason and understand the provider's data handling.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • 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
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (4)

Description-Behavior Mismatch

Medium
Confidence
90% confidence
Finding
The top-level description does not disclose that the skill supports direct proxy requests to Datagma endpoints, effectively enabling broader API interaction than the manifest suggests. Understating arbitrary request capability is risky because proxy access can reach endpoints and data flows beyond the curated action list, weakening user and policy expectations.

Description-Behavior Mismatch

Medium
Confidence
90% confidence
Finding
The top-level description does not disclose that the skill supports direct proxy requests to Datagma endpoints, effectively enabling broader API interaction than the manifest suggests. Understating arbitrary request capability is risky because proxy access can reach endpoints and data flows beyond the curated action list, weakening user and policy expectations.

Vague Triggers

Medium
Confidence
84% confidence
Finding
The activation text 'Use when the user wants to interact with Datagma data' is overly broad and may cause the skill to trigger for a wide range of generic enrichment or CRM-related requests. Over-broad routing increases the chance that sensitive person-lookup or enrichment functions are invoked unnecessarily or without sufficiently specific user intent.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The skill documents reverse email lookup, phone lookup, people search, work-email discovery, and raw proxy requests without any privacy, consent, jurisdiction, or data-handling guardrails. Because these operations involve personal data enrichment and contact discovery, omission of warnings and use constraints can facilitate privacy-invasive or noncompliant data processing.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal