Action Network

Security checks across malware telemetry and agentic risk

Overview

This is a coherent Action Network integration, but it gives an agent broad live read/write authority through Membrane without clear confirmation guardrails.

Install only if you are comfortable connecting Action Network through Membrane and giving the agent access to live organizing data. Use a least-privileged account where possible, review the exact connection, action, and JSON input before running commands, and require explicit confirmation before creating, updating, deleting, publishing, emailing, or otherwise changing records.

SkillSpector

By NVIDIA

Vulnerability Patterns

Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (2)

Vague Triggers

Medium

Confidence: 90% confidence
Finding: The skill description is broad enough to match many generic requests about managing data or automating workflows, which can cause the agent to invoke this skill in situations where the user did not clearly intend to access or modify Action Network. Because the skill exposes both read and write capabilities, overbroad routing increases the chance of unintended actions against a live external system.

Missing User Warnings

Medium

Confidence: 94% confidence
Finding: The documentation advertises management of Action Network data and workflows but does not warn that the skill can create or update records in a production activism platform. Without an explicit warning or confirmation requirement, an agent may perform destructive or irreversible changes based on ambiguous prompts or inferred intent.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal